Item 7. MANAGEMENT’S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS OF OPERATIONS
85K characters. Original on sec.gov · Markdown
Item 7. MANAGEMENT’S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS OF OPERATIONS
Overview
Favorable demographic and economic conditions, combined with historically low interest rates, have supported the recovery in U.S. new home sales that began in 2012. During this period, we have made significant investments to acquire and develop land inventory and open new communities, including opening approximately 250 new communities across our local markets in each of the last three years. We have grown our investment in the business in a disciplined manner by emphasizing smaller projects and working to shorten our years of owned land supply, including increasing the use of land option agreements, which now account for 40% of our controlled lots as compared with 11% at the beginning of 2012. We have also focused our land investments on closer-in locations where we think demand is more sustainable when the market ultimately moderates. We have accepted the trade-off of having to pay more for certain land positions where we can be more confident in future performance. The combination of favorable demand conditions, our investments in new communities, and our focus on gross margin performance through community location, strategic pricing, and construction efficiencies resulted in growth in our revenues and income before income taxes each year during the period from 2012 to 2018.
We entered 2018 with a large backlog of new orders, and demand conditions remained favorable through the early part of 2018, as evidenced by continued growth in new orders during the traditional spring selling season. However, this was followed by an industry-wide softening in demand that began in the second quarter of 2018. To varying degrees, the slowdown has occurred across all major buyer groups and all of our geographies. This slowdown was closely correlated with the rise in mortgage interest rates that began in May 2018, however, we believe that the broader cause is the affordability challenge that many prospective buyers continue to face, which has created uncertainty in the industry regarding short-term demand. However, many of the fundamentals supporting continued growth in demand, including: a strong employment picture in the U.S.; high consumer confidence; a supportive, though slightly higher, interest rate environment; and a limited supply of new and existing homes, remain favorable.
We believe that the actions we have taken over the past few years to shorten the duration of our land inventory, increase our use of land option agreements, and drive higher margins while maintaining a conservative financial position allow us to operate effectively in most economic conditions. Additionally, our overall financial condition continues to support investing in the business while returning excess capital to shareholders. If demand conditions accelerate, we have the communities and lots available to meet that demand.
The following is a summary of our operating results by line of business ($000's omitted, except per share data):
| Years Ended December 31, | |||||||||||
| 2018 | 2017 | 2016 | |||||||||
| Income before income taxes: | |||||||||||
| Homebuilding | $ | 1,288,804 | $ | 865,332 | $ | 860,766 | |||||
| Financial Services | 58,736 | 73,496 | 73,084 | ||||||||
| Income before income taxes | 1,347,540 | 938,828 | 933,850 | ||||||||
| Income tax expense | (325,517 | ) | (491,607 | ) | (331,147 | ) | |||||
| Net income | $ | 1,022,023 | $ | 447,221 | $ | 602,703 | |||||
| Per share data - assuming dilution: | |||||||||||
| Net income | $ | 3.55 | $ | 1.44 | $ | 1.75 |
| • | Homebuilding income before income taxes improved each year from 2016 to 2018. Revenues increased each year and overhead leverage improved. Homebuilding income before income taxes also reflected the following significant income (expense) items ($000's omitted): |
| 2018 | 2017 | 2016 | |||||||||||
| Land inventory impairments (see Note 2) | Home sale cost of revenues | (70,965 | ) | (88,952 | ) | (1,074 | ) | ||||||
| Warranty claim (see Note 11) | Home sale cost of revenues | — | (12,389 | ) | — | ||||||||
| Net realizable value adjustments ("NRV") - land held for sale (see Note 2) | Land sale cost of revenues | (11,489 | ) | (83,576 | ) | (1,105 | ) | ||||||
| California land sale gains (see Note 3) | Land sale revenues / cost of revenues | 26,401 | — | — | |||||||||
| Insurance reserve adjustments (see Note 11) | Selling, general, and administrative expenses | 35,873 | 97,789 | 57,132 | |||||||||
| Write-offs of insurance receivables (see Note 11) | Selling, general, and administrative expenses | — | (29,624 | ) | — | ||||||||
| Restructuring costs from corporate office relocation and other actions | Selling, general, and administrative expenses | — | — | (10,030 | ) | ||||||||
| Other expense, net | — | — | (11,643 | ) | |||||||||
| Write-offs of deposits and pre-acquisition costs (see Note 2) | Other expense, net | (16,992 | ) | (11,367 | ) | (17,157 | ) | ||||||
| Impairments of unconsolidated entities (see Note 2) | Other expense, net | — | (8,017 | ) | — | ||||||||
| Settlement of disputed land transaction (see Note 11) | Other expense, net | — | — | (15,000 | ) | ||||||||
| $ | (37,172 | ) | $ | (136,136 | ) | $ | 1,123 |
For additional information on the above, see the applicable Notes to the Consolidated Financial Statements.
| • | The decrease in Financial Services income in 2018 compared with 2017 and 2016 was primarily due to a $16.1 million increase in loan origination liabilities in 2018 (see Note 11) combined with a more competitive pricing environment. Refinance activity has slowed in the mortgage industry, which has increased competition, pressured loan pricing, and resulted in lower capture rate and reduced margins on our loan originations in 2018. These factors offset higher revenues driven primarily by higher volumes in the Homebuilding segment. |
| • | Our effective tax rate was 24.2%, 52.4%, and 35.5% for 2018, 2017, and 2016, respectively (see Note 8). The effective tax rates for 2018 and 2017 reflect the impact of the Tax Act, which lowered the federal tax rate from 35% to 21% effective in 2018. Due to the Tax Act's enactment in December 2017, income tax expense for 2017 included a charge of $172.1 million related to the remeasurement of our deferred tax balances and other effects. |
Homebuilding Operations
The following is a summary of income before income taxes for our Homebuilding operations ($000’s omitted):
| Years Ended December 31, | |||||||||||||||||
| 2018 | FY 2018 vs. FY 2017 | 2017 | FY 2017 vs. FY 2016 | 2016 | |||||||||||||
| Home sale revenues | $ | 9,818,445 | 18 | % | $ | 8,323,984 | 12 | % | $ | 7,451,315 | |||||||
| Land sale and other revenues (a) (c) | 164,504 | 167 | % | 61,542 | 40 | % | 44,089 | ||||||||||
| Total Homebuilding revenues | 9,982,949 | 19 | % | 8,385,526 | 12 | % | 7,495,404 | ||||||||||
| Home sale cost of revenues (b) | (7,540,937 | ) | 17 | % | (6,461,152 | ) | 16 | % | (5,587,974 | ) | |||||||
| Land sale cost of revenues (a) | (126,560 | ) | (6 | )% | (134,449 | ) | 319 | % | (32,115 | ) | |||||||
| Selling, general, and administrative expenses ("SG&A") (d) | (1,012,023 | ) | 14 | % | (891,581 | ) | (7 | )% | (957,150 | ) | |||||||
| Other expense, net (e) | (14,625 | ) | (56 | )% | (33,012 | ) | (42 | )% | (57,399 | ) | |||||||
| Income before income taxes | $ | 1,288,804 | 49 | % | $ | 865,332 | 1 | % | $ | 860,766 | |||||||
| Supplemental data**:** | |||||||||||||||||
| Gross margin from home sales (b) | 23.2 | % | 80 bps | 22.4 | % | (260) bps | 25.0 | % | |||||||||
| SG&A % of home sale revenues (d) | 10.3 | % | (40) bps | 10.7 | % | (210) bps | 12.8 | % | |||||||||
| Closings (units) | 23,107 | 10 | % | 21,052 | 6 | % | 19,951 | ||||||||||
| Average selling price | $ | 425 | 8 | % | $ | 395 | 6 | % | $ | 373 | |||||||
| Net new orders (f): | |||||||||||||||||
| Units | 22,833 | 1 | % | 22,626 | 11 | % | 20,326 | ||||||||||
| Dollars | $ | 9,675,529 | 3 | % | $ | 9,361,534 | 21 | % | $ | 7,753,399 | |||||||
| Cancellation rate | 14 | % | 14 | % | 15 | % | |||||||||||
| Active communities at December 31 | 815 | 3 | % | 790 | 9 | % | 726 | ||||||||||
| Backlog at December 31: | |||||||||||||||||
| Units | 8,722 | (3 | )% | 8,996 | 21 | % | 7,422 | ||||||||||
| Dollars | $ | 3,836,147 | (4 | )% | $ | 3,979,064 | 35 | % | $ | 2,941,512 |
| (a) | Includes net gains of $26.4 million related to two land sale transactions in California during the year ended December 31, 2018 (see Note 3). |
| (b) | Includes the amortization of capitalized interest; land inventory impairments of $71.0 million in 2018*,* $89.0 million in 2017*, and* $1.1 million in 2016 (see Note 2); and a warranty charge of $12.4 million related to a closed-out community in 2017 (see Note 11). |
| (c) | Includes net realizable value adjustments on land held for sale of $11.5 million*,* $83.6 million*, and* $1.1 million in 2018*,* 2017*, and* 2016*, respectively (see* Note 2). |
| (d) | Includes write-offs of $29.6 million of insurance receivables associated with the resolution of certain insurance matters in 2017 (see Note 11); insurance reserve reversals of $35.9 million*,* $97.8 million and $57.1 million in 2018*,* 2017*, and* 2016*, respectively (see* Note 11); and restructuring costs from corporate office relocation and other actions of $10.0 million in 2016*.* |
| (e) | See "Other expense, net" for a table summarizing significant items. |
| (f) | Net new orders excludes backlog acquired from Wieland in January 2016 (see Note 1). Net new order dollars represent a composite of new order dollars combined with other movements of the dollars in backlog related to cancellations and change orders. |
Home sale revenues
Home sale revenues for 2018 were higher than 2017 by $1.5 billion, or 18%. The increase was attributable to a 10% increase in closings and an 8% increase in the average selling price. The increase in closings reflects the significant land investments we have made in recent years and the resulting growth in our active communities combined with the favorable buyer demand environment that continued into the spring of 2018. The higher average selling prices occurred across the majority of our markets and reflects shifts in product mix, including a higher mix of move-up homebuyers and an increase in the mix of closings in Northern California, where our average selling prices are significantly higher than the Company average.
Home sale revenues for 2017 were higher than 2016 by $872.7 million, or 12%. The increase was attributable to a 6% increase in closings and a 6% increase in the average selling price. The increase in closings reflects the significant land investments we have made in recent years and the resulting increase in our active communities combined with favorable buyer demand conditions. The increased closings occurred despite the disruption in our operations caused by Hurricane Harvey in Houston, Texas, and Hurricane Irma in Florida, as well as permitting and other municipal approval delays in certain communities. The higher average selling price for 2017 occurred across the majority of our markets and reflected a shift toward move-up homebuyers.
Home sale gross margins
Home sale gross margins were 23.2% in 2018, compared with 22.4% in 2017 and 25.0% in 2016. Our results in 2018 and 2017 include the effect of the aforementioned land inventory impairments totaling $71.0 million and $89.0 million, respectively. Excluding such impairments, gross margins remained strong in both 2018 and 2017 relative to historical levels and reflect a combination of factors, including shifts in community mix and a small increase in the mix of closings in Northern California in 2018 partially offset by the aforementioned warranty charge of $12.4 million in 2017 related to a closed-out community in Florida and slightly higher amortized interest costs (1.8% of home sale revenues in 2018 compared with 1.7% in 2017). Gross margins decreased in 2017 compared with 2016 as the result of the aforementioned land inventory impairments and warranty charge combined with higher house construction and land costs as the supply chain responded to the housing recovery.
Land sales
We periodically elect to sell parcels of land to third parties in the event such assets no longer fit into our strategic operating plans or are zoned for commercial or other development. Land sale revenues and their related gains or losses vary between periods, depending on the timing of land sales and our strategic operating decisions. Land sales contributed net gains (losses) of $37.9 million, $(72.9) million, and $12.0 million in 2018, 2017, and 2016, respectively. The gains in 2018 resulted primarily from two land sale transactions in California that contributed $26.4 million. The losses in 2017 resulted primarily from the aforementioned net realizable value charges of $83.6 million (see Note 2).
SG&A
SG&A as a percentage of home sale revenues was 10.3% and 10.7% in 2018 and 2017, respectively. The gross dollar amount of our SG&A increased $120.4 million, or 14%, in 2018 compared with 2017. The improved overhead leverage reflects volume efficiencies and realized cost efficiencies, as well as the aforementioned insurance reserve reversals of $35.9 million and $97.8 million in 2018 and 2017, respectively, partially offset by write-offs of $29.6 million in 2017 associated with the resolution of certain insurance matters (see Note 11).
SG&A as a percentage of home sale revenues was 10.7% and 12.8% in 2017 and 2016, respectively. The gross dollar amount of our SG&A decreased $65.6 million, or 7%, in 2017 compared with 2016. SG&A includes the aforementioned insurance receivable write-offs of $29.6 million in 2017 and general liability insurance reserve reversals of $97.8 million and $57.1 million in 2017 and 2016, respectively, resulting from favorable claims experience (see Note 11). Excluding these items, the improvement in our year-over-year SG&A leverage was primarily attributable to cost efficiencies realized in late 2016 that continued into 2017.
Other expense, net
Other expense, net includes the following ($000’s omitted):
| 2018 | 2017 | 2016 | |||||||||
| Write-offs of deposits and pre-acquisition costs (Note 2) | $ | (16,992 | ) | $ | (11,367 | ) | $ | (17,157 | ) | ||
| Lease exit and related costs (a) | (240 | ) | (1,729 | ) | (11,643 | ) | |||||
| Amortization of intangible assets (Note 1) | (13,800 | ) | (13,800 | ) | (13,800 | ) | |||||
| Interest income | 7,593 | 2,537 | 3,236 | ||||||||
| Interest expense | (618 | ) | (503 | ) | (686 | ) | |||||
| Equity in earnings (loss) of unconsolidated entities (Note 4) (b) | 2,690 | (1,985 | ) | 8,337 | |||||||
| Miscellaneous, net (c) | 6,742 | (6,165 | ) | (25,686 | ) | ||||||
| Total other expense, net | $ | (14,625 | ) | $ | (33,012 | ) | $ | (57,399 | ) |
| (a) | Lease exit and related costs for 2016 resulted from actions taken to reduce overheads and the substantial completion of our corporate headquarters relocation from Michigan to Georgia, which began in 2013. |
| (b) | Includes an $8.0 million impairment of an investment in an unconsolidated entity in 2017 (see Note 2). |
| (c) | Miscellaneous, net includes a charge of $15.0 million in 2016 related to the settlement of a disputed land transaction (see Note 11). |
Net new orders
Net new orders in units increased 1% in 2018 compared with 2017. The increase resulted primarily from the higher number of active communities, which increased 3% to 815 at December 31, 2018. Net new orders in dollars increased by 3% compared with 2017 due to the growth in units combined with the higher average selling price. The cancellation rate (canceled orders for the period divided by gross new orders for the period) remained stable in 2018 at 14%. Ending backlog units, which represent orders for homes that have not yet closed, decreased 3% as measured in units and 4% as measured in dollars at December 31, 2018 compared with December 31, 2017. The higher average sales price when compared to 2017 also contributed to the higher backlog dollars. Our higher number of active communities combined with the overall demand environment resulted in a strong start to the year. However, while customer traffic to our communities increased during 2018, we experienced lower than expected conversions of traffic to signups, especially among first-time and move-up buyers, beginning in May 2018 when mortgage rates increased, which compounded existing housing affordability issues faced by many homebuyers.
Net new orders in units increased 11% in 2017 compared with 2016. The increase resulted primarily from the higher number of active communities, which increased 9% to 790 active communities at December 31, 2017. Net new orders in dollars increased by 21% compared with 2016 due to the growth in units combined with the higher average selling price. The cancellation rate (canceled orders for the period divided by gross new orders for the period) decreased in 2017 from 2016 at 14% and 15%, respectively. Ending backlog units, which represent orders for homes that have not yet closed, increased 21% at December 31, 2017 compared with December 31, 2016 as measured in units and increased 35% over the prior year period as measured in dollars. The higher average sales price when compared to 2016 also contributed to the higher backlog dollars.
Homes in production
The following is a summary of our homes in production at December 31, 2018 and 2017:
| 2018 | 2017 | |||||
| Sold | 6,245 | 6,246 | ||||
| Unsold | ||||||
| Under construction | 2,531 | 1,973 | ||||
| Completed | 715 | 637 | ||||
| 3,246 | 2,610 | |||||
| Models | 1,216 | 1,148 | ||||
| Total | 10,707 | 10,004 |
The number of homes in production at December 31, 2018 was 7% higher compared to December 31, 2017. The increase in homes under production resulted from a 24% increase in the number of unsold, or "spec", homes, which resulted primarily from the strategic decision to allow spec production to run higher than in previous periods to ensure access to construction suppliers and to position communities heading into 2019 ahead of the spring selling season.
Controlled lots
The following is a summary of our lots under control at December 31, 2018 and 2017:
| December 31, 2018 | December 31, 2017 | |||||||||||||||||
| Owned | Optioned | Controlled | Owned | Optioned | Controlled | |||||||||||||
| Northeast | 5,813 | 3,694 | 9,507 | 5,194 | 5,569 | 10,763 | ||||||||||||
| Southeast | 15,800 | 11,806 | 27,606 | 15,404 | 11,085 | 26,489 | ||||||||||||
| Florida | 18,652 | 15,855 | 34,507 | 18,458 | 11,887 | 30,345 | ||||||||||||
| Midwest | 10,097 | 11,883 | 21,980 | 10,612 | 9,196 | 19,808 | ||||||||||||
| Texas | 14,380 | 11,035 | 25,415 | 13,923 | 8,320 | 22,243 | ||||||||||||
| West | 24,788 | 5,774 | 30,562 | 25,662 | 6,099 | 31,761 | ||||||||||||
| Total | 89,530 | 60,047 | 149,577 | 89,253 | 52,156 | 141,409 | ||||||||||||
| Developed (%) | 39 | % | 21 | % | 32 | % | 37 | % | 20 | % | 31 | % |
Of our controlled lots, 89,530 and 89,253 were owned and 60,047 and 52,156 were under land option agreements at December 31, 2018 and 2017, respectively. While competition for well-positioned land is robust, we continue to pursue strategic land positions that drive appropriate returns on invested capital. The remaining purchase price under our land option agreements totaled $2.6 billion at December 31, 2018. These land option agreements generally may be canceled at our discretion and in certain cases extend over several years. Our maximum exposure related to these land option agreements is generally limited to our deposits and pre-acquisition costs, which totaled $218.6 million, of which $11.2 million is refundable, at December 31, 2018.
Homebuilding Segment Operations
Our homebuilding operations represent our core business. Homebuilding offers a broad product line to meet the needs of homebuyers in our targeted markets. As of December 31, 2018, we conducted our operations in 44 markets located throughout 24 states. For reporting purposes, our Homebuilding operations are aggregated into six reportable segments:
| Northeast: | Connecticut, Maryland, Massachusetts, New Jersey, New York, Pennsylvania, Virginia | |
| Southeast: | Georgia, North Carolina, South Carolina, Tennessee | |
| Florida: | Florida | |
| Midwest: | Illinois, Indiana, Kentucky, Michigan, Minnesota, Ohio | |
| Texas: | Texas | |
| West: | Arizona, California, Nevada, New Mexico, Washington |
We also have a reportable segment for our financial services operations, which consist principally of mortgage banking and title operations. The Financial Services segment operates generally in the same markets as the Homebuilding segments.
The following table presents selected financial information for our reportable Homebuilding segments:
| Operating Data by Segment ($000's omitted) | |||||||||||||||||
| Years Ended December 31, | |||||||||||||||||
| 2018 | FY 2018 vs. FY 2017 | 2017 | FY 2017 vs. FY 2016 | 2016 | |||||||||||||
| Home sale revenues: | |||||||||||||||||
| Northeast | $ | 795,211 | 15 | % | $ | 693,624 | — | % | $ | 696,003 | |||||||
| Southeast | 1,740,239 | 12 | % | 1,556,615 | 5 | % | 1,485,809 | ||||||||||
| Florida | 1,911,537 | 30 | % | 1,469,005 | 15 | % | 1,274,237 | ||||||||||
| Midwest | 1,492,572 | 4 | % | 1,435,692 | 16 | % | 1,233,110 | ||||||||||
| Texas | 1,296,183 | 11 | % | 1,166,843 | 13 | % | 1,033,387 | ||||||||||
| West | 2,582,703 | 29 | % | 2,002,205 | 16 | % | 1,728,769 | ||||||||||
| $ | 9,818,445 | 18 | % | $ | 8,323,984 | 12 | % | $ | 7,451,315 | ||||||||
| Income before income taxes (a)****: | |||||||||||||||||
| Northeast (b) | $ | 29,629 | 40 | % | $ | 21,190 | (74 | )% | $ | 81,991 | |||||||
| Southeast | 202,639 | 65 | % | 122,532 | (16 | )% | 145,011 | ||||||||||
| Florida (c) | 289,418 | 39 | % | 208,825 | 2 | % | 205,049 | ||||||||||
| Midwest | 179,568 | 1 | % | 178,231 | 48 | % | 120,159 | ||||||||||
| Texas | 193,946 | 6 | % | 182,862 | 20 | % | 152,355 | ||||||||||
| West (d) | 511,828 | 123 | % | 229,504 | 2 | % | 225,771 | ||||||||||
| Other homebuilding (e) | (118,224 | ) | (52 | )% | (77,812 | ) | (12 | )% | (69,570 | ) | |||||||
| $ | 1,288,804 | 49 | % | $ | 865,332 | 1 | % | $ | 860,766 | ||||||||
| Closings (units): | |||||||||||||||||
| Northeast | 1,558 | 17 | % | 1,335 | (6 | )% | 1,418 | ||||||||||
| Southeast | 4,220 | 9 | % | 3,888 | — | % | 3,901 | ||||||||||
| Florida | 4,771 | 24 | % | 3,861 | 12 | % | 3,441 | ||||||||||
| Midwest | 3,716 | 1 | % | 3,696 | 8 | % | 3,418 | ||||||||||
| Texas | 4,212 | 3 | % | 4,107 | 10 | % | 3,726 | ||||||||||
| West | 4,630 | 11 | % | 4,165 | 3 | % | 4,047 | ||||||||||
| 23,107 | 10 | % | $ | 21,052 | 6 | % | 19,951 | ||||||||||
| Average selling price: | |||||||||||||||||
| Northeast | $ | 510 | (2 | )% | $ | 520 | 6 | % | $ | 491 | |||||||
| Southeast | 412 | 3 | % | 400 | 5 | % | 381 | ||||||||||
| Florida | 401 | 6 | % | 380 | 3 | % | 370 | ||||||||||
| Midwest | 402 | 3 | % | 388 | 8 | % | 361 | ||||||||||
| Texas | 308 | 8 | % | 284 | 2 | % | 277 | ||||||||||
| West | 558 | 16 | % | 481 | 13 | % | 427 | ||||||||||
| $ | 425 | 8 | % | $ | 395 | 6 | % | $ | 373 |
| (a) | Includes land-related charges as summarized in the following land-related charges table (see Note 2). |
| (b) | Northeast includes a charge of $15.0 million in 2016 related to the settlement of a disputed land transaction (see Note 11). |
| (c) | Florida includes a warranty charge of $12.4 million in 2017 related to a closed-out community (see Note 11). |
| (d) | Includes gains of $26.4 million related to two land sale transactions in California in 2018 |
| (e) | Other homebuilding includes the amortization of intangible assets, amortization of capitalized interest, and other items not allocated to the operating segments. Also includes: write-off of $29.6 million of insurance receivables associated with the resolution of certain insurance matters in 2017*; insurance reserve reversals of* $35.9 million*,* $97.8 million and $57.1 million in 2018*,* 2017*, and* 2016*, respectively (see* Note 11); and costs associated with the relocation of our corporate headquarters totaling $8.3 million in 2016*.* |
The following tables present additional selected financial information for our reportable Homebuilding segments:
| Operating Data by Segment ($000's omitted) | ||||||||||||||||||
| Years Ended December 31, | ||||||||||||||||||
| 2018 | FY 2018 vs. FY 2017 | 2017 | FY 2017 vs. FY 2016 | 2016 | ||||||||||||||
| Net new orders - units: | ||||||||||||||||||
| Northeast | 1,516 | 4 | % | 1,460 | 7 | % | 1,361 | |||||||||||
| Southeast | 4,114 | (3 | )% | 4,233 | 11 | % | 3,810 | |||||||||||
| Florida | 4,982 | 21 | % | 4,121 | 15 | % | 3,585 | |||||||||||
| Midwest | 3,631 | (6 | )% | 3,876 | 7 | % | 3,636 | |||||||||||
| Texas | 4,278 | 4 | % | 4,121 | 9 | % | 3,793 | |||||||||||
| West | 4,312 | (10 | )% | 4,815 | 16 | % | 4,141 | |||||||||||
| 22,833 | 1 | % | 22,626 | 11 | % | 20,326 | ||||||||||||
| Net new orders - dollars: | ||||||||||||||||||
| Northeast | $ | 799,373 | 6 | % | $ | 757,679 | 12 | % | $ | 674,066 | ||||||||
| Southeast | 1,721,103 | 2 | % | 1,691,020 | 14 | % | 1,483,139 | |||||||||||
| Florida | 2,029,999 | 27 | % | 1,594,367 | 19 | % | 1,340,181 | |||||||||||
| Midwest | 1,492,453 | (2 | )% | 1,523,153 | 13 | % | 1,351,828 | |||||||||||
| Texas | 1,332,598 | 10 | % | 1,214,149 | 15 | % | 1,060,217 | |||||||||||
| West | 2,300,003 | (11 | )% | 2,581,166 | 40 | % | 1,843,968 | |||||||||||
| $ | 9,675,529 | 3 | % | $ | 9,361,534 | 21 | % | $ | 7,753,399 | |||||||||
| Cancellation rates: | ||||||||||||||||||
| Northeast | 10 | % | 12 | % | 11 | % | ||||||||||||
| Southeast | 12 | % | 12 | % | 15 | % | ||||||||||||
| Florida | 13 | % | 12 | % | 12 | % | ||||||||||||
| Midwest | 12 | % | 11 | % | 12 | % | ||||||||||||
| Texas | 19 | % | 18 | % | 18 | % | ||||||||||||
| West | 17 | % | 16 | % | 19 | % | ||||||||||||
| 14 | % | 14 | % | 15 | % | |||||||||||||
| Unit backlog: | ||||||||||||||||||
| Northeast | 470 | (8 | )% | 512 | 32 | % | 387 | |||||||||||
| Southeast | 1,610 | (6 | )% | 1,716 | 25 | % | 1,371 | |||||||||||
| Florida | 1,889 | 13 | % | 1,678 | 18 | % | 1,418 | |||||||||||
| Midwest | 1,402 | (6 | )% | 1,487 | 14 | % | 1,307 | |||||||||||
| Texas | 1,492 | 5 | % | 1,426 | 1 | % | 1,412 | |||||||||||
| West | 1,859 | (15 | )% | 2,177 | 43 | % | 1,527 | |||||||||||
| 8,722 | (3 | )% | 8,996 | 21 | % | 7,422 | ||||||||||||
| Backlog dollars: | ||||||||||||||||||
| Northeast | $ | 257,812 | 2 | % | $ | 253,650 | 34 | % | $ | 189,595 | ||||||||
| Southeast | 699,030 | (3 | )% | 718,166 | 23 | % | 583,760 | |||||||||||
| Florida | 800,051 | 17 | % | 681,589 | 23 | % | 556,226 | |||||||||||
| Midwest | 588,420 | — | % | 588,539 | 17 | % | 501,079 | |||||||||||
| Texas | 486,212 | 8 | % | 449,797 | 12 | % | 402,491 | |||||||||||
| West | 1,004,622 | (22 | )% | 1,287,323 | 82 | % | 708,361 | |||||||||||
| $ | 3,836,147 | (4 | )% | $ | 3,979,064 | 35 | % | $ | 2,941,512 |
The following table presents additional selected financial information for our reportable Homebuilding segments:
| Operating Data by Segment ($000's omitted) | ||||||||||||
| Years Ended December 31, | ||||||||||||
| 2018 | 2017 | 2016 | ||||||||||
| Land-related charges:* | ||||||||||||
| Northeast | $ | 74,488 | $ | 51,362 | $ | 2,079 | ||||||
| Southeast | 8,140 | 55,689 | 3,089 | |||||||||
| Florida | 1,166 | 9,702 | 715 | |||||||||
| Midwest | 7,361 | 8,917 | 3,383 | |||||||||
| Texas | 1,204 | 2,521 | 515 | |||||||||
| West | 5,159 | 56,995 | 8,960 | |||||||||
| Other homebuilding | 1,928 | 6,726 | 595 | |||||||||
| $ | 99,446 | $ | 191,912 | $ | 19,336 |
| *** | Land-related charges include land impairments, net realizable value adjustments for land held for sale, and write-offs of deposits and pre-acquisition costs. Other homebuilding consists primarily of write-offs of capitalized interest resulting from land-related charges. See Notes 2 and 3 to the Consolidated Financial Statements for additional discussion of these charges. |
Northeast:
The length and complexity of the entitlement process in the Northeast had led to a lack of growth in volumes in recent years, which changed in 2018 with progress in a number of communities. For 2018, Northeast home sale revenues increased 15% compared with 2017 due to a 17% increase in closings, partially offset by a 2% decrease in average selling price. The higher revenues occurred across the majority of markets, which was partially offset by our exit of the St. Louis market in 2018. The increased income before income taxes resulted from the higher revenues, partially offset by higher land-related charges and increased overhead expense. Net new orders increased slightly.
For 2017, Northeast home sale revenues remained flat compared with 2016 due to a 6% decrease in closings offset by a 6% increase in average selling price. The decrease in closings occurred in the New England and Mid-Atlantic markets, while the increase in average selling price occurred across all markets. The New England closings decrease was driven primarily by closings delayed as the result of a fire in an attached product building that was under construction. The decreased income before income taxes resulted from lower margins and increased SG&A expense across all markets, combined with the aforementioned land-related charges recognized in the period (see Note 2). Net new orders increased across all markets.
Southeast:
For 2018, Southeast home sale revenues increased 12% compared with 2017 due to a 3% increase in the average selling price combined with a 9% increase in closings. The increase in the average selling price occurred across all markets except Georgia, while closings increased in Raleigh, Charlotte and Coastal Carolinas. Income before income taxes increased primarily as a result of higher revenues and reduced land-related charges in 2018. Net new orders decreased 3%, attributable to a majority of markets.
For 2017, Southeast home sale revenues increased 5% compared with 2016 due to a 5% increase in the average selling price. The increases in the average selling price occurred across all markets except Georgia, while closings decreased in Raleigh, Charlotte and Coastal Carolinas, offset by increases in Georgia and Tennessee. Income before income taxes decreased 16% primarily due to the aforementioned land-related charges, partially offset by lower SG&A expense. Net new orders increased 11%, primarily in Georgia and Raleigh.
Florida:
For 2018, Florida home sale revenues increased 30% compared with 2017 due to a 6% increase in the average selling price combined with a 24% increase in closings. The increased income before income taxes for 2018 resulted primarily from higher revenues combined with the aforementioned $12.4 million warranty charge in 2017 related to a closed-out community. Net new orders increased 21% in 2018.
For 2017, Florida home sale revenues increased 15% compared with 2016 due to a 3% increase in the average selling price combined with a 12% increase in closings. The increased income before income taxes for 2017 resulted primarily from higher revenues, partially offset by the aforementioned $12.4 million warranty charge in 2017 related to a closed-out community. Net new orders increased by 15% in 2017. Both closings and new orders increased despite the disruption in our operations caused by Hurricane Irma.
Midwest:
For 2018, Midwest home sale revenues increased 4% compared with the prior year period due to an 1% increase in closings combined with an 3% increase in the average selling price. The higher revenues occurred across the majority markets, partially offset by our decision to exit the St. Louis market in 2017, which we completed in 2018. Income before income taxes remained consistent with the prior year due to the increased revenues, partially offset by lower margins and higher SG&A expense. Net new orders decreased across substantially all markets.
For 2017, Midwest home sale revenues increased 16% compared with the prior year period due to a 8% increase in closings combined with a 8% increase in the average selling price. The higher revenues and increased closings occurred across all markets. The increased closing volume combined with lower SG&A expense led to a 48% increase in income before income taxes. Net new orders increased across all markets except for St. Louis, where we announced our decision to exit the market.
Texas:
For 2018, Texas home sale revenues increased 11% compared with the prior year period due to a 3% increase in closings combined with an 8% increase in the average selling price. The increase in average selling price occurred across all markets, while the increase in closings occurred across all markets except for Dallas and San Antonio. The higher revenues and higher closings led to increased income before income taxes. Net new orders increased 4% across all markets except for Houston which remained flat compared with 2017.
For 2017, Texas home sale revenues increased 13% compared with the prior year period due to a 10% increase in closings combined with an 2% increase in the average selling price. The increase in average selling price occurred primarily in Central Texas and San Antonio, while the increase in closings occurred across all markets except for San Antonio. The higher revenues and higher closings led to increased income before income taxes. Net new orders increased 9% across all markets except for San Antonio. Both closings and new orders increased despite the disruption in our Houston operations caused by Hurricane Harvey.
West:
For 2018, West home sale revenues increased 29% compared with the prior year period due to an 11% increase in closings combined with a 16% increase in the average selling price. The increased revenues occurred across substantially all markets but were driven primarily by Northern California. The increased revenues contributed to increased income before income taxes in all markets except New Mexico, with the majority coming from Northern California. Income before income taxes also benefited from two land sale transactions that resulted in gains totaling $26.4 million as well as the lower land-related charges. Net new orders decreased by 10% in 2018 compared with 2017, which was primarily concentrated in Northern California.
For 2017, West home sale revenues increased 16% compared with the prior year period due to a 3% increase in closings combined with a 13% increase in the average selling price. The increased closings primarily occurred in Southern California, offset by a decrease in Northern California due to permitting and other municipal approval delays in certain communities. The increased average selling price occurred across all markets. Income before income taxes slightly increased due to the increased revenues and reduced overheads, partially offset by the aforementioned land-related charges recognized during the period (see Note 2). Net new orders increased by 16% in 2017 compared with 2016 due to higher order levels across all markets.
Financial Services Operations
We conduct our Financial Services operations, which include mortgage banking, title, and insurance brokerage operations, through Pulte Mortgage and other subsidiaries. In originating mortgage loans, we initially use our own funds, including funds available pursuant to credit agreements with third parties. Substantially all of the loans we originate are sold in the secondary market within a short period of time after origination, generally within 30 days. We also sell the servicing rights for the loans we originate through fixed price servicing sales contracts to reduce the risks and costs inherent in servicing loans. This strategy results in owning the loans and related servicing rights for only a short period of time. Operating as a captive business model primarily targeted to supporting our Homebuilding operations, the business levels of our Financial Services operations are highly correlated to Homebuilding. Our Homebuilding customers continue to account for substantially all loan production. We believe that our capture rate, which represents loan originations from our Homebuilding operations as a percentage of total loan opportunities from our Homebuilding operations, excluding cash closings, is an important metric in evaluating the effectiveness of our captive mortgage business model. The following table presents selected financial information for our Financial Services operations ($000’s omitted):
| Years Ended December 31, | |||||||||||||||||
| 2018 | FY 2018 vs. FY 2017 | 2017 | FY 2017 vs. FY 2016 | 2016 | |||||||||||||
| Mortgage operations revenues | $ | 149,642 | 2 | % | $ | 146,358 | 3 | % | $ | 142,262 | |||||||
| Title and insurance brokerage revenues | 55,740 | 22 | % | 45,802 | 18 | % | 38,864 | ||||||||||
| Total Financial Services revenues | 205,382 | 7 | % | 192,160 | 6 | % | 181,126 | ||||||||||
| Expenses | (147,422 | ) | 24 | % | (119,289 | ) | 10 | % | (108,573 | ) | |||||||
| Other income, net | 776 | 24 | % | 625 | 18 | % | 531 | ||||||||||
| Income before income taxes | $ | 58,736 | (20 | )% | $ | 73,496 | 1 | % | $ | 73,084 | |||||||
| Total originations: | |||||||||||||||||
| Loans | 14,464 | 2 | % | 14,152 | 6 | % | 13,373 | ||||||||||
| Principal | $ | 4,456,360 | 8 | % | $ | 4,127,084 | 11 | % | $ | 3,706,745 |
| Years Ended December 31, | |||||||||||
| 2018 | 2017 | 2016 | |||||||||
| Supplemental data: | |||||||||||
| Capture rate | 76.2 | % | 79.9 | % | 81.2 | % | |||||
| Average FICO score | 752 | 749 | 750 | ||||||||
| Loan application backlog | $ | 2,012,340 | $ | 2,263,803 | $ | 1,670,160 | |||||
| Funded origination breakdown: | |||||||||||
| Government (FHA, VA, USDA) | 20 | % | 22 | % | 23 | % | |||||
| Other agency | 68 | % | 70 | % | 70 | % | |||||
| Total agency | 88 | % | 92 | % | 93 | % | |||||
| Non-agency | 12 | % | 8 | % | 7 | % | |||||
| Total funded originations | 100 | % | 100 | % | 100 | % |
Revenues
Total Financial Services revenues during 2018 increased 7% compared with 2017. The increase is primarily due to higher loan origination, title, and insurance brokerage volume resulting from higher volumes in the Homebuilding segment. A higher average loan size, driven primarily by higher average selling prices in the Homebuilding segment, also contributed to the higher revenues. These factors were partially offset by the lower capture rate resulting from a more competitive market environment. Total Financial Services revenues during 2017 increased 6% compared with 2016 due to higher mortgage and title volumes resulting from increased home closings in the Homebuilding segment, partially offset by lower mortgage revenue per loan, which were largely attributable to increased competition and pressured loan pricing.
Income before income taxes
The decrease in income before income taxes for 2018 as compared with 2017 was primarily due to a $16.1 million increase in loan origination liabilities in 2018 (see Note 11) combined with a more competitive pricing environment. Refinance activity has slowed in the mortgage industry, which has increased competition, pressured loan pricing, and resulted in lower margins on our loan originations in 2018. These factors offset higher revenues driven primarily by higher volumes in the Homebuilding segment. The increased income before income taxes for 2017 as compared with 2016 resulted from higher origination volume and an increase in the revenue per loan combined with better overhead leverage and contributions from our title operations.
Income Taxes
Our effective tax rate was 24.2%, 52.4% and 35.5% for 2018, 2017, and 2016, respectively. The effective tax rates for 2018 and 2017 reflect the impact of the Tax Act, which lowered the federal tax rate from 35% to 21% effective in 2018. Due to the Tax Act's enactment in December 2017, income tax expense for 2017 included a charge of $172.1 million related to the remeasurement of our deferred tax balances and other effects. The effective tax rate for 2016 included a net benefit related to the reversal of a portion of our valuation allowance related to a legal entity restructuring along with the resolution of certain state income tax and other matters.
Liquidity and Capital Resources
We finance our land acquisition, development, and construction activities and financial services operations using internally-generated funds supplemented by credit arrangements with third parties and capital market financing. We routinely monitor current and expected operational requirements and financial market conditions to evaluate accessing other available financing sources, including revolving bank credit and securities offerings.
At December 31, 2018, we had unrestricted cash and equivalents of $1.1 billion, restricted cash balances of $23.6 million, and $760.6 million available under our revolving credit facility. We follow a diversified investment approach for our cash and equivalents by maintaining such funds with a broad portfolio of banks within our group of relationship banks in high quality, highly liquid, short-term deposits and investments.
We retired outstanding debt totaling $82.8 million, $134.7 million, and $986.9 million during 2018, 2017, and 2016, respectively. Our ratio of debt-to-total capitalization, excluding our Financial Services debt, was 38.6%, which is within our targeted range of 30.0% to 40.0%, at December 31, 2018.
Unsecured senior notes
In February 2016, we issued $1.0 billion of unsecured senior notes, consisting of $300.0 million of 4.25% senior notes due March 1, 2021, and $700.0 million of 5.50% senior notes due March 1, 2026. The net proceeds from this senior notes issuance were used to fund the retirement of $465.2 million of our senior notes that matured in May 2016, with the remaining net proceeds used for general corporate purposes. In July 2016, we issued an additional $1.0 billion of unsecured senior notes, consisting of $400.0 million of 4.25% senior notes due March 1, 2021, and $600.0 million of 5% senior notes due January 15, 2027. The net proceeds from the July senior notes issuance were used for general corporate purposes and to pay down approximately $500.0 million of outstanding debt, including the remainder of a then existing term loan facility. At December 31, 2018, we had $3.0 billion of unsecured senior notes outstanding with no repayments due until March 2021 when $700.0 million of notes are scheduled to mature.
Revolving credit facility
In June 2018, we entered into the Second Amended and Restated Credit Agreement ("Revolving Credit Facility") which replaced the Company's previous credit agreement. The Revolving Credit Facility contains substantially similar terms to the previous credit agreement and extended the maturity date from June 2019 to June 2023. The Revolving Credit Facility has a maximum borrowing capacity of $1.0 billion and contains an uncommitted accordion feature that could increase the capacity to $1.5 billion, subject to certain conditions and availability of additional bank commitments. The Revolving Credit Facility also provides for the issuance of letters of credit that reduce the available borrowing capacity under the Revolving Credit Facility, with a sublimit of $500.0 million at December 31, 2018. The interest rate on borrowings under the Revolving Credit Facility may be based on either the London Interbank Offered Rate ("LIBOR") or a base rate plus an applicable margin, as defined
therein. We had no borrowings outstanding and $239.4 million and $235.5 million of letters of credit issued under the Revolving Credit Facility at December 31, 2018 and 2017, respectively.
The Revolving Credit Facility contains financial covenants that require us to maintain a minimum Tangible Net Worth, a minimum Interest Coverage Ratio, and a maximum Debt-to-Capitalization Ratio (as each term is defined in the Revolving Credit Facility). As of December 31, 2018, we were in compliance with all covenants. Outstanding balances under the Revolving Credit Facility are guaranteed by certain of our wholly-owned subsidiaries. Our available and unused borrowings under the Revolving Credit Facility, net of outstanding letters of credit, amounted to $760.6 million and $764.5 million as of December 31, 2018 and 2017, respectively.
Other notes payable
Certain of our local homebuilding operations are party to non-recourse and limited recourse collateralized notes payable with third parties that totaled $41.3 million at December 31, 2018. These notes have maturities ranging up to three years, are secured by the applicable land positions to which they relate, have no recourse to any other assets, and are classified within notes payable.
Pulte Mortgage
Pulte Mortgage provides mortgage financing for the majority of our home closings by utilizing its own funds and funds made available pursuant to credit agreements with third parties. Pulte Mortgage uses these resources to finance its lending activities until the loans are sold in the secondary market, which generally occurs within 30 days.
Pulte Mortgage maintains a master repurchase agreement with third party lenders. In August 2018, Pulte Mortgage entered into an amended and restated repurchase agreement (the “Repurchase Agreement”) that extended the termination date to August 2019. The maximum aggregate commitment was $520.0 million during the seasonally high borrowing period from December 26, 2018 through January 14, 2019. At all other times, the maximum aggregate commitment ranges from $240.0 million to $400.0 million. The purpose of the changes in capacity during the term of the agreement is to lower associated fees during seasonally lower volume periods of mortgage origination activity. Borrowings under the Repurchase Agreement are secured by residential mortgage loans available-for-sale. The Repurchase Agreement contains various affirmative and negative covenants applicable to Pulte Mortgage, including quantitative thresholds related to net worth, net income, and liquidity. Pulte Mortgage had $348.4 million and $437.8 million outstanding under the Repurchase Agreement at December 31, 2018, and 2017, respectively, and was in compliance with its covenants and requirements as of such dates.
Share repurchase program
In 2013, our Board of Directors authorized and announced a share repurchase program, which was subsequently increased by $1.0 billion in July 2016 and by $500.0 million in January 2018. We repurchased 10.9 million, 35.4 million, and 30.9 million shares in 2018, 2017, and 2016, respectively, for a total of $294.6 million, $910.3 million, and $600.0 million in 2018, 2017, and 2016, respectively, under this program. At December 31, 2018, we had remaining authorization to repurchase $299.9 million of common shares.
Dividends
Our declared quarterly cash dividends totaled $108.5 million, $110.0 million, and $122.2 million in 2018, 2017, and 2016, respectively.
Cash flows
Operating activities
Our net cash provided by operating activities in 2018 was $1.4 billion, compared with net cash provided by operating activities of $663.1 million and $68.3 million in 2017 and 2016, respectively. Generally, the primary drivers of our cash flow from operations are profitability and changes in inventory levels and residential mortgage loans available-for-sale. Our positive cash flow from operations for 2018 was primarily due to our net income of $1.0 billion, which included non-cash land-related charges of $99.4 million and $362.8 million of deferred income tax expense, supplemented by a $107.3 million reduction in residential mortgage loans available-for-sale. These factors were partially offset by a net increase in inventories of $50.4 million resulting from higher levels of spec inventory.
Our positive cash flow from operations for 2017 was primarily due to our net income of $447.2 million, which included $191.9 million in non-cash land-related charges and deferred tax expense of $422.3 million. These were partially offset by a net increase in inventories of $569.0 million resulting from ongoing land acquisition and development investment to support future growth combined with additional house inventory to support the higher backlog.
Our negative cash flow from operations for 2016 was primarily due to a net increase in inventories of $897.1 million resulting from increased land investment, combined with a net increase in residential mortgage loans available-for-sale of $99.5 million.
Investing activities
Net cash used in investing activities totaled $41.9 million in 2018, compared with $50.2 million in 2017 and $471.2 million in 2016. The use of cash from investing activities in 2018 was primarily due to $59.0 million of capital expenditures, which increased from 2017 as the result of new community openings combined with increased expenditures on information technology solutions. The use of cash from investing activities in 2017 was primarily due to $32.1 million of capital expenditures and $23.0 million for investments in unconsolidated subsidiaries. The use of cash from investing activities in 2016 was primarily due to the acquisition of certain real estate assets from Wieland (see Note 1).
Financing activities
Net cash used in financing activities was $580.3 million in 2018, compared with $1.0 billion during 2017 and net cash provided by financing activities of $350.7 million in 2016. The net cash used in financing activities for 2018 resulted primarily from the repurchase of 10.9 million common shares for $294.6 million under our repurchase authorization, repayments of debt of $82.8 million, cash dividends of $104.0 million, and net repayments of $89.4 million under the Repurchase Agreement related to the aforementioned decrease in residential mortgage loans available-for-sale.
Net cash used in financing activities for 2017 resulted primarily from the repurchase of 35.4 million common shares for $910.3 million under our repurchase authorization, repayments of debt of $134.7 million, and cash dividends of $112.7 million, partially offset by net borrowings of $106.2 million under the Repurchase Agreement related to a seasonal increase in residential mortgage loans available-for-sale. Cash provided by financing activities for 2016 resulted primarily from the proceeds of the unsecured senior notes issuance for $2.0 billion, offset by the repayment of $986.9 million of debt and the repurchase of 30.9 million common shares for $600.0 million under our repurchase authorization and cash dividends of $124.7 million.
Inflation
We, and the homebuilding industry in general, may be adversely affected during periods of inflation because of higher land and construction costs. Inflation may also increase our financing costs. In addition, higher mortgage interest rates affect the affordability of our products to prospective homebuyers. While we attempt to pass on to our customers increases in our costs through increased sales prices, market forces may limit our ability to do so. If we are unable to raise sales prices enough to compensate for higher costs, or if mortgage interest rates increase significantly, our revenues, gross margins, and net income could be adversely affected.
Seasonality
Although significant changes in market conditions have impacted our seasonal patterns in the past and could do so again, we historically experience variability in our quarterly results from operations due to the seasonal nature of the homebuilding industry. We generally experience increases in revenues and cash flow from operations during the fourth quarter based on the timing of home closings. This seasonal activity increases our working capital requirements in our third and fourth quarters to support our home production and loan origination volumes. As a result of the seasonality of our operations, our quarterly results of operations are not necessarily indicative of the results that may be expected for the full year.
Contractual Obligations and Commercial Commitments
The following table summarizes our payments under contractual obligations as of December 31, 2018:
| Payments Due by Period ($000’s omitted) | |||||||||||||||||||
| Total | 2019 | 2020-2021 | 2022-2023 | After 2023 | |||||||||||||||
| Contractual obligations: | |||||||||||||||||||
| Notes payable (a) | $ | 4,582,517 | $ | 191,379 | $ | 1,034,534 | $ | 271,250 | $ | 3,085,354 | |||||||||
| Operating lease obligations | 113,496 | 24,806 | 35,553 | 27,269 | 25,868 | ||||||||||||||
| Total contractual obligations (b) | $ | 4,696,013 | $ | 216,185 | $ | 1,070,087 | $ | 298,519 | $ | 3,111,222 |
| (a) | Represents principal and interest payments related to our senior notes and limited recourse collateralized financing arrangements. |
| (b) | We do not have any payments due in connection with capital lease or long-term purchase obligations. |
We are subject to certain obligations associated with entering into contracts (including land option contracts) for the purchase, development, and sale of real estate in the routine conduct of our business. Option contracts for the purchase of land enable us to defer acquiring portions of properties owned by third parties and unconsolidated entities until we have determined whether to exercise our option, which may serve to reduce our financial risks associated with long-term land holdings. At December 31, 2018, we had $218.6 million of deposits and pre-acquisition costs, of which $11.2 million is refundable, relating to option agreements to acquire 60,047 lots with a remaining purchase price of $2.6 billion. We expect to acquire the majority of such land within the next three years.
We are currently under examination by various taxing jurisdictions and anticipate finalizing the examinations with certain jurisdictions within the next twelve months. The final outcome of these examinations is not yet determinable. The statute of limitations for our major tax jurisdictions remains open for examination for tax years January 1, 2005 - January 1, 2018. At December 31, 2018, we had $30.6 million of gross unrecognized tax benefits and $5.8 million of related accrued interest and penalties.
The following table summarizes our other commercial commitments as of December 31, 2018:
| Amount of Commitment Expiration by Period ($000’s omitted) | |||||||||||||||||||
| Total | 2019 | 2020-2021 | 2022-2023 | After 2023 | |||||||||||||||
| Other commercial commitments: | |||||||||||||||||||
| Guarantor credit facilities (a) | $ | 1,000,000 | $ | — | $ | — | $ | 1,000,000 | $ | — | |||||||||
| Non-guarantor credit facilities (b) | 520,000 | 520,000 | — | — | — | ||||||||||||||
| Total commercial commitments (c) | $ | 1,520,000 | $ | 520,000 | $ | — | $ | 1,000,000 | $ | — |
| (a) | The $1.0 billion in 2022-2023 represents the capacity of our unsecured revolving credit facility, under which no borrowings were outstanding, and $239.4 million of letters of credit were issued at December 31, 2018*.* |
| (b) | Represents the capacity of the Repurchase Agreement, of which $348.4 million was outstanding at December 31, 2018*. The capacity of* $520.0 million was effective through January 14, 2019 after which it ranges from $240.0 million to $400.0 million until its expiration in August 2019*.* |
| (c) | The above table excludes an aggregate $1.3 billion of surety bonds, which typically do not have stated expiration dates. |
Off-Balance Sheet Arrangements
We use letters of credit and surety bonds to guarantee our performance under various contracts, principally in connection with the development of our homebuilding projects. The expiration dates of the letter of credit contracts coincide with the expected completion date of the related homebuilding projects. If the obligations related to a project are ongoing, annual extensions of the letters of credit are typically granted on a year-to-year basis. At December 31, 2018, we had outstanding letters of credit of $239.4 million. Our surety bonds generally do not have stated expiration dates; rather, we are released from
the bonds as the contractual performance is completed. These bonds, which approximated $1.3 billion at December 31, 2018, are typically outstanding over a period of approximately three to five years. Because significant construction and development work has been performed related to the applicable projects but has not yet received final acceptance by the respective counterparties, the aggregate amount of surety bonds outstanding is in excess of the projected cost of the remaining work to be performed.
In the ordinary course of business, we enter into land option agreements in order to procure land for the construction of houses in the future. At December 31, 2018, these agreements had an aggregate remaining purchase price of $2.6 billion. Pursuant to these land option agreements, we provide a deposit to the seller as consideration for the right to purchase land at different times in the future, usually at predetermined prices.
At December 31, 2018, aggregate outstanding debt of unconsolidated joint ventures was $42.9 million, of which $42.1 million was related to one joint venture in which we have a 50% interest. In connection with this loan, we and our joint venture partner provided customary limited recourse guaranties in which our maximum financial loss exposure is limited to our pro rata share of the debt outstanding. See Note 4 to the Consolidated Financial Statements for additional information.
Critical Accounting Policies and Estimates
The accompanying consolidated financial statements were prepared in conformity with U.S. generally accepted accounting principles. When more than one accounting principle, or the method of its application, is generally accepted, we select the principle or method that is appropriate in our specific circumstances (see Note 1 to our Consolidated Financial Statements). Application of these accounting principles requires us to make estimates about the future resolution of existing uncertainties; as a result, actual results could differ from these estimates. In preparing these consolidated financial statements, we have made our best estimates and judgments of the amounts and disclosures included in the consolidated financial statements, giving due regard to materiality.
Revenue recognition
Home sale revenues - Home sale revenues and related profit are generally recognized when title to and possession of the home are transferred to the buyer at the home closing date. Little to no estimation is involved in recognizing such revenues.
Land sale revenues - We periodically elect to sell parcels of land to third parties in the event such assets no longer fit into our strategic operating plans or are zoned for commercial or other development. Land sales are generally outright sales of specified land parcels with cash consideration due on the closing date, which is generally when performance obligations are satisfied. Certain land sale contracts may contain unique terms that require management judgment in determining the appropriate revenue recognition, but the impact of such transactions is generally immaterial.
Financial services revenues - Loan origination fees, commitment fees, and certain direct loan origination costs are recognized as incurred. Expected gains and losses from the sale of residential mortgage loans and their related servicing rights are included in the measurement of written loan commitments that are accounted for at fair value through Financial Services revenues at the time of commitment. The determination of fair value for certain of these financial instruments requires the use of estimates and management judgment. Subsequent changes in the fair value of these loans are reflected in Financial Services revenues as they occur. Interest income is accrued from the date a mortgage loan is originated until the loan is sold. Mortgage servicing fees represent fees earned for servicing loans for various investors. Servicing fees are based on a contractual percentage of the outstanding principal balance, or a contracted set fee in the case of certain sub-servicing arrangements, and are credited to income when related mortgage payments are received or the sub-servicing fees are earned.
Revenues associated with our title operations are recognized as closing services are rendered and title insurance policies are issued, both of which generally occur as each home is closed. Insurance brokerage commissions relate to commissions on home and other insurance policies placed with third party carriers through various agency channels. Our performance obligations for policy renewal commissions are considered satisfied upon issuance of the initial policy, and related contract assets for estimated future renewal commissions are included in other assets and totaled $30.8 million at December 31, 2018. Due to uncertainties in the estimation process and the long duration of renewal policies, which can extend years into the future, actual results could differ from such estimates.
Inventory and cost of revenues
Inventory is stated at cost unless the carrying value is determined to not be recoverable, in which case the affected inventory is written down to fair value. Cost includes land acquisition, land development, and home construction costs, including interest, real estate taxes, and certain direct and indirect overhead costs related to development and construction. For those communities for which construction and development activities have been idled, applicable interest and real estate taxes are expensed as incurred. Land acquisition and development costs are allocated to individual lots using an average lot cost determined based on the total expected land acquisition and development costs and the total expected home closings for the community. The specific identification method is used to accumulate home construction costs.
We capitalize interest cost into homebuilding inventories. Each layer of capitalized interest is amortized over a period that approximates the average life of communities under development. Interest expense is allocated over the period based on the timing of home closings.
Cost of revenues includes the construction cost, average lot cost, estimated warranty costs, and closing costs applicable to the home. Sales commissions are classified within selling, general, and administrative expenses. The construction cost of the home includes amounts paid through the closing date of the home, plus an accrual for costs incurred but not yet paid, based on an analysis of budgeted construction costs. This accrual is reviewed for accuracy based on actual payments made after closing compared with the amount accrued, and adjustments are made if needed. Total community land acquisition and development
costs are based on an analysis of budgeted costs compared with actual costs incurred to date and estimates to complete. The development cycles for our communities range from under one year to in excess of ten years for certain master planned communities. Adjustments to estimated total land acquisition and development costs for the community affect the amounts costed for the community’s remaining lots.
We test inventory for impairment when events and circumstances indicate that the undiscounted cash flows estimated to be generated by the community may be less than its carrying amount. Such indicators include gross margins or sales paces significantly below expectations, construction costs or land development costs significantly in excess of budgeted amounts, significant delays or changes in the planned development for the community, and other known qualitative factors. Communities that demonstrate potential impairment indicators are tested for impairment by comparing the expected undiscounted cash flows for the community to its carrying value. For those communities whose carrying values exceed the expected undiscounted cash flows, we determine the fair value of the community and impairment charges are recorded if the fair value of the community’s inventory is less than its carrying value.
We generally determine the fair value of each community using a combination of discounted cash flow models and market comparable transactions, where available. These estimated cash flows are significantly impacted by estimates related to expected average selling prices and sale incentives, expected sales paces, expected land development and construction timelines, and anticipated land development, construction, and overhead costs. The assumptions used in the discounted cash flow models are specific to each community. Due to uncertainties in the estimation process, the significant volatility in demand for new housing, the long life cycles of many communities, and potential changes in our strategy related to certain communities, actual results could differ significantly from such estimates.
Residential mortgage loans available-for-sale
In accordance with ASC 825, “Financial Instruments” (“ASC 825”), we use the fair value option for our residential mortgage loans available-for-sale. Election of the fair value option for residential mortgage loans available-for-sale allows a better offset of the changes in fair values of the loans and the derivative instruments used to economically hedge them without having to apply complex hedge accounting provisions. Changes in the fair value of these loans are reflected in revenues as they occur.
Loan origination liabilities
Our mortgage operations may be responsible for losses associated with mortgage loans originated and sold to investors in the event of errors or omissions relating to representations and warranties made by us that the loans met certain requirements, including representations as to underwriting standards, the existence of primary mortgage insurance, and the validity of certain borrower representations in connection with the loan. If a loan is determined to be faulty, we either indemnify the investor for potential future losses, repurchase the loan from the investor, or reimburse the investor's actual losses. Estimating the required liability for these potential losses requires a significant level of management judgment. Given the unsettled litigation, changes in values of underlying collateral over time, and other uncertainties regarding the ultimate resolution of these claims, actual costs could differ from our current estimates.
Allowance for warranties
Home purchasers are provided with a limited warranty against certain building defects, including a one-year comprehensive limited warranty and coverage for certain other aspects of the home’s construction and operating systems for periods of up to (and in limited instances exceeding) 10 years. We estimate the costs to be incurred under these warranties and record a liability in the amount of such costs at the time revenue is recognized. Factors that affect our warranty liability include the number of homes sold, historical and anticipated rates of warranty claims, and the projected cost of claims. We periodically assess the adequacy of our recorded warranty liability for each geographic market in which we operate and adjust the amounts as necessary. Actual warranty costs in the future could differ from our estimates.
Income taxes
We evaluate our deferred tax assets each period to determine if a valuation allowance is required based on whether it is "more likely than not" that some portion of the deferred tax assets would not be realized. The ultimate realization of these deferred tax assets is dependent upon the generation of sufficient taxable income during future periods. We conduct our evaluation by considering all available positive and negative evidence. This evaluation considers, among other factors, historical operating results, forecasts of future profitability, the duration of statutory carryforward periods, and the outlooks for
the U.S. housing industry and broader economy. The accounting for deferred taxes is based upon estimates of future results. Differences between estimated and actual results could result in changes in the valuation of our deferred tax assets that could have a material impact on our consolidated results of operations or financial position. Changes in existing tax laws could also affect actual tax results and the realization of deferred tax assets over time.
Unrecognized tax benefits represent the difference between tax positions taken or expected to be taken in a tax return and the benefits recognized for financial statement purposes. We follow the provisions of ASC 740, “Income Taxes” (“ASC 740”), which prescribes a minimum recognition threshold a tax position is required to meet before being recognized in the financial statements. Significant judgment is required to evaluate uncertain tax positions. Our evaluations of tax positions consider a variety of factors, including relevant facts and circumstances, applicable tax law, correspondence with taxing authorities, and effective settlements of audit issues. Changes in the recognition or measurement of uncertain tax positions could result in material increases or decreases in income tax expense (benefit) in the period in which the change is made. Interest and penalties related to income taxes and unrecognized tax benefits are recognized as a component of income tax expense (benefit).
Self-insured risks
At any point in time, we are managing over 1,000 individual claims related to general liability, property, errors and omission, workers compensation, and other business insurance coverage. We reserve for costs associated with such claims (including expected claims management expenses) on an undiscounted basis at the time product revenue is recognized for each home closing and periodically evaluate the recorded liabilities based on actuarial analyses of our historical claims. The actuarial analyses calculate estimates of the ultimate cost of all unpaid losses, including estimates for incurred but not reported losses ("IBNR"). IBNR represents losses related to claims incurred but not yet reported plus development on reported claims.
Our recorded reserves for all such claims totaled $737.0 million and $758.8 million at December 31, 2018 and 2017, respectively, the vast majority of which relate to general liability claims. The recorded reserves include loss estimates related to both (i) existing claims and related claim expenses and (ii) IBNR and related claim expenses. Liabilities related to IBNR and related claim expenses represented approximately 65% of the total general liability reserves at December 31, 2018 and 2017. The actuarial analyses that determine the IBNR portion of reserves consider a variety of factors, including the frequency and severity of losses, which are based on our historical claims experience supplemented by industry data. The actuarial analyses of the reserves also consider historical third party recovery rates and claims management expenses. Because of the inherent uncertainty in estimating future losses related to these claims, actual costs could differ significantly from estimated costs. Based on the actuarial analyses performed, we believe the range of reasonably possible losses related to these claims is $625 million to $850 million. While this range represents our best estimate of our ultimate liability related to these claims, due to a variety of factors, including those factors described above, there can be no assurance that the ultimate costs realized by us will fall within this range.
Housing market conditions have been volatile across most of our markets over the past ten years, and we believe such conditions can affect the frequency and cost of construction defect claims. Additionally, IBNR estimates comprise the majority of our liability and are subject to a high degree of uncertainty due to a variety of factors, including changes in claims reporting and resolution patterns, third party recoveries, insurance industry practices, the regulatory environment, and legal precedent. State regulations vary, but construction defect claims are reported and resolved over an extended period often exceeding ten years. Changes in the frequency and timing of reported claims and estimates of specific claim values can impact the underlying inputs and trends utilized in the actuarial analyses, which could have a material impact on the recorded reserves. Additionally, the amount of insurance coverage available for each policy period also impacts our recorded reserves. Because of the inherent uncertainty in estimating future losses and the timing of such losses related to these claims, actual costs could differ significantly from estimated costs.
Adjustments to reserves are recorded in the period in which the change in estimate occurs. During 2018, 2017, and 2016, we reduced general liability reserves by $35.9 million, $97.8 million, and $57.1 million, respectively, as a result of changes in estimates resulting from actual claim experience observed being less than anticipated in previous actuarial projections. The changes in actuarial estimates were driven by changes in actual claims experience that, in turn, impacted actuarial estimates for potential future claims. These changes in actuarial estimates did not involve any changes in actuarial methodology but did impact the development of estimates for future periods, which resulted in adjustments to the IBNR portion of our recorded liabilities.
In certain instances, we have the ability to recover a portion of our costs under various insurance policies or from subcontractors or other third parties. Estimates of such amounts are recorded when recovery is considered probable. Our receivables from insurance carriers totaled $153.0 million and $213.4 million at December 31, 2018 and 2017, respectively. The
insurance receivables relate to costs incurred or to be incurred to perform corrective repairs, settle claims with customers, and other costs related to the continued progression of both known and anticipated future construction defect claims that we believe to be insured related to previously closed homes. We believe collection of these insurance receivables is probable based on various factors, including, the legal merits of our positions after review by legal counsel, favorable legal rulings received to date, the credit quality of our carriers, and our long history of collecting significant amounts of insurance reimbursements under similar insurance policies related to similar claims, including significant amounts funded by the above carriers under different policies.
While the outcome of these matters cannot be predicted with certainty, we do not believe that the resolution of such matters will have a material adverse impact on our results of operations, financial position, or cash flows.
Previous: Item 6. SELECTED FINANCIAL DATA · Next: Item 7A. QUANTITATIVE AND QUALITATIVE DISCLOSURES ABOUT MARKET RISK