![](https://capedge.com/proxy/CORRESP/0000950103-11-000892/logo.jpg)
March 4, 2011 |
Mr. Rufus Decker
Accounting Branch Chief
Division of Corporate Finance
Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, DC 20549
Re: | Gafisa S.A. (“Gafisa” or the “Company”) |
Form 20-F for the Fiscal Year Ended December 31, 2009 | |
Filed on March 10, 2010 | |
File No. 001-33356 |
Dear Mr. Decker:
Set forth below is Gafisa’s response to your letter dated February 18, 2011 relating to Gafisa’s Form 20-F for the year ended December 31, 2009 (the “Form 20-F”). To assist in the Staff’s review of the responses, each response is preceded with the text (in bold type) of the comment as stated in your letter.
Capitalized terms used in the responses set forth below and not otherwise defined herein have the meanings set forth in the Form 20-F.
General
1. | Where a comment below requests additional disclosures or other revisions to be made, please show us in your supplemental response what the revisions will look like. These revisions should be included in your future filings. Some of our comments refer to US GAAP literature. If your accounting under Brazilian GAAP differs from your accounting under US GAAP, please also show us the additional disclosures that will be included in your US GAAP reconciliation footnote. |
The Company acknowledges the Staff’s comment and will provide additional disclosures or revisions in future filings, as appropriate, as set forth below.
Furthermore, the Company intends to revise the US GAAP footnote of its historical financial statements to retrospectively account for certain financial instruments as
Mr. Rufus Decker | 2 | March 4, 2011 |
marketable securities instead of as cash equivalents and to otherwise correct any material matters in response to the Staff’s comments.
Item 3 – Information on the Company
Overview of Residential Buildings, page 25
2. | We note your response to comment two from our letter dated November 23, 2010. However, it appears that your table summarizing projects completed in 2009 or prior where the number of units sold as of December 31, 2009 was less than 90% is incomplete. We noted several projects in your disclosures on pages 25 to 33 which appear to meet the specified criteria noted above but were excluded from the table in your response letter. For example, your disclosures on pages 25 to 33 indicate that Residencial Renata, Residencial Bela Vista, Arsenal Life, Belford Roxo Garden, Santo Antonio Life, and Alphaville – Campo Grande were all completed during 2009 or prior and had less than 90% of their units sold as of December 31, 2009. This is not intended to be an all inclusive list. Please revisit the completeness of your prior response and provide us with an update as necessary for any additional projects which were completed in 2009 or prior but had less than 90% of their units sold as of December 31, 2009. For each of these projects, please similarly explain the circumstances surrounding the number of unsold units and explain how you evaluated these projects for impairment as of December 31, 2009. |
In response to the Staff’s comments, below please find an updated table:
Project | Year completed | Percentage of units sold disclosed in Form 20-F | Actual percentage of units sold as of December 31, 2009 | |||
Newly disclosed projects: | ||||||
Residencial Bela Vista | 2008 | 87.0% | 97.0% | |||
Arsenal Life | 2008 | —% | 98.4% | |||
Alphaville Campo Grande | 2009 | 81.0% | 99.0% | |||
Residencial Trabalhador | 2008 | —% | 87.0% | |||
Residencial Trabalhador – Phase 2 | 2008 | —% | 87.0% | |||
Projects previously disclosed: | ||||||
Bairro Novo – Phase 3 | 2010 | —% | 99.6% | |||
Cotia – Phase 4 – Stage 1 | 2010 | —% | 91.7% | |||
Cotia – Phase 4 – Stage 2 | 2010 | —% | 74.6% | |||
Residencial Monaco | 2012 | —% | 100.0% | |||
Residencial Monte Cristo I (a) | 2010 | —% | 6.0% | |||
Fit Coqueiro | 2010 | —% | 97.9% | |||
Fit Mirante do Lago(b) | 2011 | —% | 32.2% | |||
Novo Jockey 1 | 2010 | —% | 100.0% | |||
Novo Jockey 2 | 2011 | —% | 100.0% | |||
Viver Melhor | 2011 | —% | 100.0% | |||
Camaçari Ville I –Residencial Jardim Atlantico Life (c) | 2011 | —% | 18.6% | |||
Camaçari Ville II – Residencial Camaçari Life (d) | 2011 | —% | 34.4% | |||
Camaçari Ville III – Residencial Camaçari Duo | 2011 | —% | 97.0% |
In response to the Staff’s comment, the Company confirms that it will correct prior disclosures that appear in any future filings.
The following projects were not yet completed as of December 31, 2009 and where the number of units sold as of December 31, 2009 was less than 50%.
Mr. Rufus Decker | 3 | March 4, 2011 |
a) | Residencial Monte Cristo I: |
This development was 5.5% completed at December 31, 2009 and the only cost is related to the land; the development was canceled in 2010. The Company is performing studies to launch another development in the near future (on or about 2013) at the same location. The Company has no reason to believe that the carrying value of this land is below its market value.
b) | Fit Mirante do Lago: |
This development was 12.1% completed at December 31, 2009 and only 32.2% of the units have been sold. In order to accelerate the sales of units of this development, the Company opened a sales office in the city of Belem. As a result, currently, 91.1% of the units have been sold — all of them above cost.
c) | Camaçari Ville I – Residencial Jardim Atlântico Life: |
This development was 8.9% completed at December 31, 2009 and only 18.6% of the units have been sold. According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 60.8% of the development was completed and 83.3% of the units were sold.
d) | Camaçari Ville II – Residencial Camaçari Life: |
This development was 37.1% completed at December 31, 2009 and only 34.4% of the units have been sold. According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 95.5% of the development was completed and 94.3% of the units were sold.
In response to the Staff’s comment, the table below discloses the projects that were completed in 2009 or in prior years and where the number of units sold as of December 31, 2009 was less than 90%. For each of the projects listed below, the Company has quantified the number of units unsold, the sales value of units unsold and the total number of completed units associated with each project:
Project | Percentage of units sold | Units unsold | Sales value of units unsold in reais | Total number of units | |||||||||
Newly disclosed projects: | |||||||||||||
Santo Antonio Life (e) | 3.1 | % | 31 | R$2,235,100 (R$72,100 each) | 32 | ||||||||
Belford Roxo Garden (f) | 12.3 | % | 533 | R$44,318,950 (R$83,150 each) | 608 | ||||||||
Residencial Trabalhador (g) | 87.0 | % | 13 | R$494,000 (R$38,000 each) | 100 | ||||||||
Residencial Trabalhador – Phase 2 (g) | 87.0 | % | 67 | R$2,546,000 (R$38,000 each) | 515 | ||||||||
Projects previously disclosed: | |||||||||||||
Belle Vue (h) | 88.9 | % | 2 | R$1,332,000 (R$666,000 each) | 18 |
Mr. Rufus Decker | 4 | March 4, 2011 |
Project | Percentage of units sold | Units unsold | Sales value of units unsold in reais | Total number of units | |||||||||
Espacio Laguna (i) | 88.8 | % | 9 | R$5,337,000 (R$593,000 each) | 80 | ||||||||
Quinta Imperial (j) | 88.3 | % | 15 | R$2,280,000 (R$152,000 each) | 128 | ||||||||
Alphaville Gravatai (k) | 83.0 | % | 111 | R$5,328,000 (R48,000 each) | 654 | ||||||||
Residencial Valencia (l) | 80.0 | % | 16 | R$ 1,376,000 (R$ 86,000 each) | 80 | ||||||||
Residencial Portal de Valença Garden (m) | 22.1 | % | 151 | R$ 9,664,000 (R$ 64,000 each) | 194 | ||||||||
Residencial Bartolomeu de Gusmão (n) | 79.2 | % | 54 | R$ 3,132,000 (R$ 58,000 each) | 260 | ||||||||
Residencial Vale do Sol (o) | 16.3 | % | 67 | R$ 3,149,000 (R$ 47,000 each) | 80 | ||||||||
Residencial Vitória Régia (p) | 40.6 | % | 38 | R$ 1,748,000 (R$ 46,000 each) | 64 |
In response to the Staff’s comment, the Company confirms that it will correct prior disclosures that appear in any future filings.
In response to the Staff’s comment, the Company provides explanations for those projects that were completed in 2009 or in prior years and where the number of units sold as of December 31, 2009 was less than 90%.
e) | Santo Antonio Life |
The development Santo Antonio Life was 95% completed at December 31, 2009 and yet only 3.1% of the units were sold. The larger percentage of unsold units is primarily due to pending legal documentation (i.e., occupancy certificates) from the development’s partner. Once this issue is resolved, the remaining units will be sold, based on the Company’s internal assessment. As of December 31, 2010, the percentage of units sold was 3.1%. The Company has concluded that impairment does not exist because this is a resolvable documentation matter and the units are anticipated to be sold at a profit.
f) | Belford Roxo Garden |
This development was 4.6% completed at December 31, 2009 and related only to the cost of land. The development was canceled in 2010 due to changes in Tenda’s product portfolio. The Company expects to sell this landbank without incurring a loss based on third party studies of the land and its location. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
g) | Residencial Trabalhador (Phases 1 and 2) |
This development was 100.0% completed at December 31, 2009 and only 87.0% of the units were sold. According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short
Mr. Rufus Decker | 5 | March 4, 2011 |
time period with sales value higher than the accounted cost. As of December 31, 2010, 100.0% of the units were sold.
h) | Belle Vue |
According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 94.4% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
i) | Espacio Laguna |
According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 97.5% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
j) | Quinta Imperial |
According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 100% of the units were sold.
k) | Alphaville Gravataí |
Most of unsold inventory units are commercial and multi-family units. These units are easier to sell after completion of the residential area because it creates a demand for services. As of December 31, 2010, 91.6% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
l) | Residencial Valencia |
According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 97.5% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
m) | Residencial Portal de Valença Garden |
The development was 54.8% completed at December 31, 2009 and yet only 22.1% of the units were sold and are delivered. This project originally had 194 units and during 2010 the Company decided to build 70 units, of which 45 (or 64.3%) had been sold as of December 31, 2010. For the remaining area of the landbank, the Company has decided to subdivide this landbank and launch a new
Mr. Rufus Decker | 6 | March 4, 2011 |
development at the same location. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
n) | Residencial Bartolomeu de Gusmão |
According to the then Company’s business plan, this development’s selling forecast indicated sales of the unsold units within a short time period with sales value higher than the accounted cost. As of December 31, 2010, 99.2% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value.
o) | Residencial Vale do Sol |
As mentioned in our response letter dated February 14, 2011, this development was 90.7% completed at December 31, 2009 and yet only 16.3% of the units were sold. The larger percentage of unsold units is primarily attributable to the use of dry-wall in the construction for which it is not possible for a customer to obtain third party financing from a Brazilian financial institution. As a result, the Company began offering financing to select customers for up to 220 months in an effort to increase sales of such units. As of December 31, 2010, 92.7% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value. Please see the Company’s response to comment 5 below.
p) | Residencial Vitória Regia |
As mentioned in our response letter dated February 14, 2011, this development was 100% completed at December 31, 2009 and yet only 40.6% of the units were sold. The larger percentage of unsold units is primarily attributable to the use of dry-wall in the construction for which it is not possible for a customer to obtain third party financing from a Brazilian financial institution. As a result, the Company began offering financing to select customers of up to 220 months in an effort to increase sales of such units. As of December 31, 2010, 83.3% of the units were sold. The Company currently has no reason to believe that the carrying value of this property is greater than its market value. Please see the Company’s response to comment 5 below.
In its February 14, 2010 response letter to the Staff, the Company described its impairment policy. To increase transparency about application of its impairment policy, the Company notes that it uses its business plan as the basis to monitor potential impairment of its inventory of condominiums under construction and held for sale. In that context, the business plan for underdevelopment properties is evaluated quarterly and an impairment analysis is performed if an indication of impairment exists based on a review of the business plan. The business plan considers all costs to complete the project as well as the expected selling price.
Furthermore, the Company has as procedure for the financial viabilities of the projects to be initiated, the approval by the Board, in which the main economic and financial indicators are the present value of cash flows (NPV) and internal rate of return (IRR)
Mr. Rufus Decker | 7 | March 4, 2011 |
of the resulting financial flows. The viabilities reflect the values of sales revenues and the costs of construction, land, commission, marketing and others, plus the speed of their realization. Specifically for sales evolution, the Company considers the sale of units over the forecast period, typically using 30% at the beginning, 30% within 6 months, 20% during construction and 20% at the conclusion of the construction.
No additional analysis is performed if the project is either selling or expected to sell at a positive margin. No projects were identified that did not have a positive margin in 2007, 2008 or 2009.
For unsold units, ongoing sales are monitored to identify break even or negative margins. Impairment analysis is performed only for properties that are selling at break even or negative margins. No such projects were identified in 2007, 2008 or 2009. At December 31, 2009, the Company had 1,488 units in its unsold inventory and for the year ended December 31, 2009, the Company sold 22,012 units.
In light of the recent strong economic environment in Brazil, the Company believes its approach is prudent and sufficiently detailed to identify any impairment. In connection with its responses to the Staff’s comments, the Company performed a look back analysis, in particular as it relates to unsold units, and identified no additional procedures.
3. | As related matter, please confirm that when you prepared your response to comment two from our letter dated November 23, 2010, you considered any projects that may have been launched prior to 2007 and therefore did not appear in your disclosures on pages 25 to 33. |
The Company confirms that the response to comment two in response to the Staff’s comment letter dated November 23, 2010 considered projects launched prior to 2007.
4. | We noted several instances in which projects have been launched but your disclosures on pages 25 to 33 provide no indication of usable area, no indication a completion year, and only a small percentage of units were sold as of December 31, 2009. For example, Aroeira Garden, Vila Riviera/Vila Positano – Phase 1, Hamburgo Garden, Munique Garden, and Santa Rita Life, This is not intended to be an all inclusive list. For each of these projects, please explain the circumstances surrounding the lack of sale and explain how you evaluated these projects for impairment as of December 31, 2009. |
In response to the Staff’s comment, below please find the revised table:
Project | Year Launched | Gafisa Participation (%) | Usable Area (m2) (1) (2) | Completion Year | Number of Units (2) | Units Sold (%) (as of December 31, 2009) | ||||||||||||||||
Vila Real Life - Sitio Cia | 2009 | 100 | % | 10,603.00 | 2011 | 178 | 99.4 | % | ||||||||||||||
Bairro Novo – Phase III | 2009 | 100 | % | 26,111.00 | 2010 | 448 | 99.6 | % | ||||||||||||||
Diamond (a) | 2009 | 100 | % | 11,927.00 | Canceled | 312 | 7.0 | % | ||||||||||||||
Cotia – Phase IV – Stage I | 2009 | 100 | % | 4,256.00 | 2010 | 96 | 91.7 | % | ||||||||||||||
Clube Garden – Mônaco | 2009 | 100 | % | 11,441.00 | 2011 | 186 | 100.0 | % | ||||||||||||||
Cotia – Phase IV – Stage II | 2009 | 100 | % | 9,930.00 | 2010 | 224 | 74.6 | % | ||||||||||||||
Portal do Sol Life I | 2009 | 100 | % | 2,354.00 | 2012 | 64 | 66.0 | % | ||||||||||||||
Portal do Sol Life II | 2009 | 100 | % | 2,354.00 | 2012 | 64 | 67.0 | % | ||||||||||||||
Portal do Sol Life III (j) | 2009 | 100 | % | 2,354.00 | 2012 | 64 | 41.0 | % | ||||||||||||||
Residencial Monet II (Grand Ville das Artes – Phase III) | 2009 | 100 | % | 4,937.00 | 2011 | 120 | 76.0 | % |
Mr. Rufus Decker | 8 | March 4, 2011 |
Project | Year Launched | Gafisa Participation (%) | Usable Area (m2) (1) (2) | Completion Year | Number of Units (2) | Units Sold (%) (as of December 31, 2009) | ||||||||||||||||
Residencial Mogi Das Cruzes Life (b) | 2008 | 100 | % | 16,009.00 | Canceled | 351 | 13.1 | % | ||||||||||||||
Residencial Itaim Paulista Life I | 2008 | 100 | % | 1,165.67 | 2011 | 160 | 100.0 | % | ||||||||||||||
Residencial Bunkyo (c) | 2008 | 100 | % | 13,944.00 | Canceled | 332 | 2.1 | % | ||||||||||||||
Residencial Portal Do Sol (j) | 2008 | 100 | % | 16,889.00 | 2012 | 282 | 30.0 | % | ||||||||||||||
Residencial Recanto Dos Passaros I (d) | 2008 | 100 | % | 9,134.00 | Canceled | 200 | 1.5 | % | ||||||||||||||
Residencial Monaco | 2008 | 100 | % | 1,384.23 | 2012 | 233 | 100.0 | % | ||||||||||||||
Residencial Monte Cristo I (l) | 2008 | 100 | % | 4,597.00 | 2010 | 96 | 6.0 | % | ||||||||||||||
Residencial Renata (j) | 2008 | 100 | % | 9,031.00 | 2011 | 200 | 5.0 | % | ||||||||||||||
Residencial Bela Vista | 2008 | 100 | % | 4,242.00 | 2008 | 101 | 97.0 | % | ||||||||||||||
Residencial Estrela Nova 1 (e) | 2008 | 100 | % | 21,544.00 | Canceled | 432 | 14.8 | % | ||||||||||||||
Parque Montebello Life I (f) | 2008 | 100 | % | 15,241.00 | Canceled | 256 | 5.5 | % | ||||||||||||||
Residencial Villa Bella | 2008 | 100 | % | 801.00 | 2009 | 16 | 100.0 | % | ||||||||||||||
Nova Marica Life (j) | 2008 | 100 | % | 21,603.00 | 2012 | 468 | 50.0 | % | ||||||||||||||
Madureira Tower (h) | 2008 | 100 | % | 4,346.00 | Canceled | 144 | — | % | ||||||||||||||
Fit Coqueiro - Stake Acquisition | 2008 | 20 | % | 35,614.00 | 2010 | 570 | 97.8 | % | ||||||||||||||
Fit Mirante do Lago – Phase I (k) | 2008 | 70 | % | 33,947.00 | 2011 | 461 | 32.2 | % | ||||||||||||||
Arsenal Life | 2007 | 100 | % | 6,819.00 | 2008 | 481 | 98.4 | % | ||||||||||||||
Vila Riviera / Vila Positano – Phase I | 2007 | 100 | % | 4,092.00 | 2009 | 84 | 95.0 | % | ||||||||||||||
Aroeira Garden (h) | 2007 | 100 | % | 5,664.00 | Canceled | 120 | - | % | ||||||||||||||
Belford Roxo Garden (h) | 2007 | 100 | % | 10,723.02 | Canceled | 608 | 12.3 | % | ||||||||||||||
Hamburgo Garden (h) | 2007 | 100 | % | 6,910.00 | Canceled | 162 | 11.7 | % | ||||||||||||||
Munique Garden (h) | 2007 | 100 | % | 5,800.00 | Canceled | 136 | 27.9 | % | ||||||||||||||
Santa Rita Life (h) | 2007 | 100 | % | 6,024.00 | Canceled | 112 | 0.9 | % | ||||||||||||||
Novo Jockey Life I | 2007 | 100 | % | 25,390.00- | 2009 | 500 | 100.0 | % | ||||||||||||||
Novo Jockey Life II | 2007 | 100 | % | 8,951.00 | 2009 | 180 | 100.0 | % | ||||||||||||||
Residencial Jardim dos Girassóis | 2007 | 100 | % | 2,746.00 | 2009 | 60 | 95.0 | % | ||||||||||||||
Residencial San Marino II | 2007 | 100 | % | 2,927.00 | 2009 | 60 | 100.0 | % | ||||||||||||||
Residencial Vale Nevado (m) | 2007 | 100 | % | 2,209.00 | Canceled | 46 | 97.8 | % | ||||||||||||||
Residencial Montana | 2007 | 100 | % | 5,524.00 | 2009 | 104 | 100.0 | % | ||||||||||||||
Residencial Tulipas (h) | 2007 | 100 | % | 6,302.00 | Canceled | 118 | 1.7 | % | ||||||||||||||
Residencial Luiz Inácio (h) | 2007 | 100 | % | 5,083.00 | Canceled | 124 | 33.1 | % | ||||||||||||||
Residencial Doze de Outubro (h) | 2007 | 100 | % | 5,739.00 | Canceled | 140 | 12.1 | % | ||||||||||||||
Residencial Santa Julia | 2007 | 100 | % | 10,041.00 | 2011 | 260 | 99.6 | % | ||||||||||||||
Residencial Filadélphia | 2007 | 100 | % | 3,497.00 | 2011 | 160 | 11.9 | % | ||||||||||||||
Viver Melhor | 2007 | 100 | % | 4,200.00 | 2011 | 100 | 100.0 | % | ||||||||||||||
Residencial Tapajos | 2007 | 100 | % | 3,980.00 | 2008 | 64 | 98.4 | % | ||||||||||||||
Residencial Parque Das Amoras | 2007 | 100 | % | 13,080.00 | 2008 | 195 | 99.0 | % | ||||||||||||||
Residencial Venda Nova Life | 2007 | 100 | % | 1,507.00 | 2008 | 34 | 100.0 | % | ||||||||||||||
Residencial Millenium (g) | 2007 | 100 | % | 10,552.00 | Canceled | 201 | 51.2 | % | ||||||||||||||
Res. Amsterdam | 2007 | 100 | % | 2,272.00 | 2008 | 48 | 100.0 | % | ||||||||||||||
Res do Trabalhador | 2007 | 100 | % | 108.10 | 2008 | 100 | 87.0 | % | ||||||||||||||
Res do Trabalhador – Phase II | 2007 | 100 | % | 1,297.20 | 2008 | 515 | 87.0 | % | ||||||||||||||
Parque Nova Esperança Life | 2007 | 100 | % | 5,165.00 | 2008 | 124 | 100.0 | % | ||||||||||||||
Portal de Valença (i) | 2007 | 100 | % | 8,055.00 | 2009 | 194 | 22.1 | % | ||||||||||||||
Camaçari Ville I – Residencial Jardim Atlantico Life (j) | 2007 | 100 | % | 28,576.00 | 2011 | 608 | 18.6 | % | ||||||||||||||
Camaçari Ville II – Residencial Camaçari Life (j) | 2007 | 100 | % | 27,025.00 | 2011 | 575 | 34.4 | % | ||||||||||||||
Camaçari Ville III – Residencial Camaçari Duo | 2007 | 100 | % | 21,808.00 | 2011 | 464 | 97.0 | % |
The following projects were canceled or had less than 50% of its units sold as of December 31, 2009:
a) | Diamond: |
This project was canceled and during 2010, a new development, Villagio Carioca, was launched at the same location. As of December 31, 2010, 19.2% of the development was completed and 39.9% of its units were sold.
The following projects were in their initial construction phase as of December 31, 2009, with only the costs related to land, and were canceled during 2010. The Company is performing studies to launch new developments in the same location in the near future (2011 and 2012).
b) | Residencial Mogi das Cruzes: |
Mr. Rufus Decker | 9 | March 4, 2011 |
The development involves the construction of buildings in an area containing protected woodlands. As a result, the new development must be approved by local governmental authorities.
c) | Residencial Bunkyo: |
This development presented a low sales velocity and the new development uses molds, which will turn the sales value feasible to the market value.
d) | Residencial Recanto dos Passaros: |
The development was not in compliance with the permanent preservation area (APP). As a result, a new development must be approved.
e) | Residencial Estrela Nova I: |
This development had a low sales velocity and this landbank is currently subject to new legislation and must be re-approved.
f) | Parque Montebello Life I: |
This development was not in compliance with the requirements relating to the water path and therefore, a new development must be approved.
g) | Residencial Millenium: |
This development had environmental issues due to the interference along the river path close to the landbank, and therefore, a new development must be approved in accordance with the permanent preservation area (APP).
Based on its business plan evaluation, the Company believes that these new developments ((b) to (g)) will be sold with no losses. The Company anticipates it will be able to secure the necessary approvals with undue delay or expense based on local requirements and standards. The Company currently has no reason to believe that the carrying value of these properties is greater than the market value.
h) | In addition, the Company intends to sell its landbank with no losses based on studies performed by land sales experts for the following developments: Hamburgo Garden, Munique Garden, Santa Rita Life, Residencial Tulipas, Residencial Luiz Inacio, Residencial Doze de Outubro, Madureira Tower, Aroeira Garden and Belford Roxo Garden. |
i) | Residencial Portal de Valença Garden: |
Please refer to item (l) in comment 2 above.
j) | Although the following developments had a low percentage of units sold as of December 31, 2009, they were under construction and have a normal sales velocity: Portal do Sol III, Residencial Portal do Sol, Residencial Renata, Nova |
Mr. Rufus Decker | 10 | March 4, 2011 |
Marica Life, Camaçari Ville I – Residencial Jardim Atlantico Life and Camaçari Ville II – Residencial Camaçari Life. According to the then Company’s business plan, the Company did not identify any instance of impairment for these developments.
k) | Fit Mirante do Lago: |
Please refer to item (b) in comment 2 above.
l) | Residencial Monte Cristo I: |
Please refer to item (a) in comment 2 above.
m) | Residencial Vale Nevado: |
This development was canceled due to certain technical issues in the project implementation and its technical feasibility. As it was a project with a low total amount of sales value, the Company decided to cancel the development. The Company intends to sell its landbank with no losses based on studies performed by land sales experts.
In its February 14, 2010 response letter to the Staff, the Company described its impairment policy. To increase transparency about application of its impairment policy, the Company notes that it uses its business plan as the basis to monitor potential impairment of its inventory of landbank properties. In that context, the business plan for the landbank properties is evaluated regularly and an impairment analysis is performed if an indication of impairment exists based on review of the business plan. The business plans considers the value of the property based on research performed by land prospection experts. In light of the recent strong economic environment in Brazil, the Company believes its approach is prudent and sufficiently detailed to identify any impairment. In connection with its response to the Staff’s comments, the Company performed look back analysis, in particular as it relates to unsold properties and identified no additional procedures. |
5. | Please tell us why the drywall technology you used for Residencial Vale do Sol and Residencial Vitoria Regia has resulted in customers not being able to obtain third party financing from a Brazilian financial institution. Please tell us the extent to which you have experienced or are experiencing this issue with any other projects. |
The Company respectfully submits that it has not experienced any workmanship, health-concern or any other similar issues with respect to the use of drywall technology. Instead, the use of this technology in certain developments for low income has resulted in customers not being able to obtain third party financing from a specific Brazilian government financial institution (Caixa Econômica Federal - CEF). CEF provides a large amount of financing, especially in the low income segment. Drywall does not meet the material construction requirements established by the CEF to provide financing. CEF requires the use of block/brick construction. The two developments noted in the comment are the only ones in which the drywall
Mr. Rufus Decker | 11 | March 4, 2011 |
technology was used, and the Company believes that it will ultimately be able to sell the remaining units. As noted in its response letter dated February 14, 2011, the Company has been offering financing to select customers of up to 220 months in an effort to increase sales of such units.
As of December 31, 2010, 92.7% and 83.3% of the units at Residencial Vale do Sol and Residencial Vitoria Regia, respectively, were sold. After taking into account sales prices and customer collections, no indication of impairment has been identified. The Company intends to monitor these for impairment under ASC 360, which addresses impairment for long-lived assets.
6. | Please tell us and revise your future filings to disclose the level at which you evaluate your properties held for sale for impairment under U.S.GAAP. For example, it is unclear if you evaluate impairment for each individual project, by neighborhood, by city or some other method. |
The Company considers impairment at the level of each individual project, regardless of segment, geographic location or stage of completion.
As noted in the February 14, 2011 response letter, the Company has evaluated its developments for impairment based on review of the Company’s business plan as described in response to comment 2 and 4, which was based on assumptions considering the positive economic trends from Brazilian economy, credit availability provided by the government, market growth for the segment and our internal selling and commercial perspectives. Based on that, the Company has not identified potential instances of impairment among its properties held for sale and, consequently, no impairments provisions have been recorded on any of its developments during 2007, 2008 and 2009. If and to the extent the Company identifies material impairment in its future filings, it will fully disclose details of the impairments and write-offs.
Customer Financing, page 38
7. | We note your response to comment three from our letter dated November 23, 2010. Please confirm that the table in your supplemental response will be part of the additional disclosure you will include in future filings. Please also clarify whether the table shows the full sales value of the terminations or just the net amount after factoring in penalties. We await your consideration of the default rates and termination practices and their implication on your U.S GAAP accounting policies for revenue recognition under the percentage of completion method for each period presented in your December 31, 2009 Form 20-F, as well as for the year ended December 31, 2010. Please provide us with the analysis you performed supporting the conclusions you reach. |
The Company confirms that the tabular information included in its February 14, 2011 response will be included in its future filings. The Company respectfully submits that such tabular information includes all cancelations, including those related to (i) Company performance issues, (ii) cancelations in which an initial investment had yet to be made by the customer and (iii) cancelations for which default was subsequently
Mr. Rufus Decker | 12 | March 4, 2011 |
agreed by the Company and a partial refund was negotiated based in part on the contractual provisions.
Furthermore, the Company confirms that tabular information shows the full sales value of terminations. Given the historic application of the percentage of completion revenue recognition, such amounts naturally are not indicative of the revenues that had been recognized under such contracts through the date of cancelation. This will also be highlighted in future filings.
As indicated in the Company’s February 14, 2011 response, both Gafisa and Tenda sales contracts are irrevocable under Brazilian law. That means that a customer does not have a unilateral ability to terminate a contract once it is executed, nor does the customer have an ability to require a refund of amounts previously paid unless the Company agrees. To the extent that a customer is not in compliance with its obligations under a contract, the Company has at its option the ability to either force compliance through the Brazilian courts, or agree to a “default” by the customer. Should the Company agree at its sole discretion to refund part of amounts paid to the defaulting party, it will normally apply the penalty guideline in the contract. Gafisa has made such refunds in the past for commercial reasons (it is in its interest to avoid protracted legal discussions and be able to quickly place the unit again on the market; with the rise in prices in recent years, the Company has consistently made gains upon remarketing the unit as well as retaining a substantial portion of the installments paid by the defaulting party) and to protect its brand. As discussed below, while Gafisa considers contractual provisions in its initial and continuing investment tests, it does not consider any amounts commercially negotiated and repaid to the defaulting party in excess of contractual amounts (inclusive of penalty provisions) to be refunds for the purpose of its initial and continuing investment tests as the contract does not provide the defaulting party with a right to cancel.
Tenda contracts provide for a refund to customers of 80% of the amounts previously paid during the construction period to the extent that default is agreed. Gafisa contracts provide for a penalty to be charged to the customer which generally equates to 18%1 of sales prices (adjusted for inflation) to the extent default is agreed. After charging such penalties, should amounts previously paid by customers under a Gafisa contract be in excess of the penalty computation, then a refund of 60% of such remaining balance is provided to the Gafisa customer based on the terms of the contract. Both Gafisa and Tenda have historically entered into commercial negotiations with customers. Tenda refunds have approximated the contractual amounts, while Gafisa has historically been willing to negotiate higher levels of refunds in an attempt to avoid protracted court proceedings and regain clear title to its units so they can be sold timely. Gafisa’s negotiated repayments2 have historically approximately 35% of amounts previously paid.
EITF Issue 06-08 “Applicability of the Assessment of a Buyer's Continuing Investment under FASB Statement No. 66 for Sales of Condominiums” required amounts potentially refundable to customers to be excluded from the initial and continuing
__________________
1 The 18% is set forth in the Gafisa contract and was rounded to 20% in our prior response.
Mr. Rufus Decker | 13 | March 4, 2011 |
investment test under ASC 360.20.40. The status of the Company’s analysis of these requirements is below.
Gafisa Segment
The Gafisa segment contracts do not provide for refunds upon default/cancelation until the customer has made payments aggregating more than 18% of the sales price (adjusted for inflation) because of the aforementioned cancelation penalties. Deposits collected from customers up to 18% of the sales price are contractually protected from refundability. Accordingly, the initial investment tests of 5% (primary residence) and 10% (secondary residency) are met and the excess is applied to the continuing investment test. Thus, the aforementioned “penalties” have no impact on the initial investment test although the Company performs the initial investment test contract-by-contract.
In considering the continuing investment test, Gafisa notes that historically its customers normally make payments of 10% of total sales price by the time foundation is complete (the first time that percentage of completion revenue recognition could be applied), and customer payments typically accumulate to 30% by the time construction is complete based on a series of scheduled payments. Gafisa projects on average take three years to complete, generally two years from the time foundation is completed. Customers meeting such parameters will normally pass the continuing investment test presuming they stay current on their periodic payment requirements3. The Company is in the process of verifying this conclusion.
Tenda Segment
Approximately 92% and 100% of projects at December 31, 2009 and 2008, respectively, were subject to Tenda’s “pilot contracts” with the remaining 8% of 2009 contracts being subject to Tenda’s “My-Home-My-Life” (“MHML”) contracts offered through Caixa Econômica Federal – CEF, a Brazilian government entity.
Tenda’s historical policy has been for a down payment of 10% to be received in order for the initial investment test to be met. Given that Tenda pilot contracts have potential refund provisions of 80% of amounts previously paid before tax, those potentially refundable amounts are excluded from the initial investment test. When those amounts are excluded, a Tenda pilot customer would be required to make an initial investment of about 50%4 of sales price before revenue could be recognized. Because low income home buyers do not make 50% down payments, Tenda pilot contracts will not meet the “initial investment test” and thus should be recognized using “deposit accounting” (based on when the unit is delivered). This required adjustment is discussed below.
Tenda’s MHML contracts do not contain a refund clause. Tenda is in the process of completing its initial and continuing investment test for MHML contracts.
Upon the completion of its quantification, the Company will make a determination as to whether its reconciliation to US GAAP consolidated net income for 2009 and 2008,
__________________
3 Based on a 30-year mortgage, which is the typical period in Brazil.
Mr. Rufus Decker | 14 | March 4, 2011 |
and reconciliation of shareholders’ equity as of December 31, 2009 and 2008 (as reconciled in its 2009 Form 20-F) will need to be amended. The Company believes it is reasonably possible such reconciliation will need to be amended. The Company is working diligently and will complete the process as soon as possible.
8. | Please revise your filings to quantify the amount of penalties each segment charged during the periods presented to customers that defaulted on their sales contracts. Please also disclose where you classified these penalties in your financial statements. To the extent that these penalties had a significant impact upon your consolidated or segment results from operations, this should be clearly identified and discussed in your MD&A. |
We respectfully refer the Staff to the Company’s response to comment seven, specifically as it relates to the adjustment to Tenda’s historical revenue recognition. As explained in that response, upon recognition of a default, the Company typically enters into commercial negotiations with its customer to determine the amount ultimately refunded. Refunds have historically averaged 76% of amounts paid for Tenda and 35% of amounts paid for Gafisa. Consistent with those historical percentages, during the years ended December 31, 2009, 2008 and 2007, penalties charged customers that have defaulted (and contracts cancelled) have been as follows: |
Company | 2009 | 2008 | 2007 |
Gafisa | R$ 5.4 million | R$2,7 | R$5.3 million |
Tenda | R$ 7.8 million | R$ 4.8 million | — |
This table includes Tenda from the date of its acquisition in October 2008. These penalties are charged in order to reimburse the Company for sales and marketing costs previously incurred, and thus they are charged as a reduction of operating revenue.
The Company will disclose in future filings the amount of penalties that each segment charged during the periods presented to the customers that defaulted on their sales contracts, and where such contracts were then cancelled. The Company will also disclose where it classified these penalties in its consolidated financial statements. To the extent these penalties had a significant impact upon the Company consolidated or segment results from operations, this will be identified and discussed in the Company’s MD&A in future filings. However, the Company currently believes that such amounts and trends will not have a material effect.
Item 5 – Operating and Financial Review and Prospects
Business Segments, page 66
9. | We note your response to comment four from our letter dated November 23, 2010. To the extent that fluctuations in the number of units planned but not launched, the amount of land held for development, or the value of unsold inventory have significantly impacted your segment results from operations, we believe that an analysis of these factors would provide greater transparency to investors about your business through the eyes of management. Please revise your MD&A to explain any significant fluctuations in the above items, as applicable. Please also revise to explain in greater detail how the mix of the types of developments under construction has changed for each of your segments during the periods presented |
Mr. Rufus Decker | 15 | March 4, 2011 |
and specifically how the change in mix affected each segment’s results of operations.
While the Company acknowledges the Staff’s request, the Company does not believe that any of the factors noted in the comment (such as fluctuations in units planned but not launched, the amount of land held for development, and the value of unsold inventory) have significantly impacted its financial position or segment results from operations during any of the periods presented.
Projects typically takes one to three years from land acquisition to being ready to launch. This lead-time is why the Company has approximately three years of “land-bank”, measured by the potential volume of future sales and is aligned with market trends. By the end of 2008, our total consolidated land-bank measured by the potential volume of future sales, reached R$ 17.84 billion, compared to R$ 15.82 billion at the end of 2009. This 11% reduction does not affect the results of our operations and, in the Company’s view, also does not affect the Company’s results, since it still represents more than three years of current launches, based on both 2008 and 2009 launches (2008 R$ 4.20 billion and 2009 R$ 2.30 billion). The Company also does not believe that any past delays in unit launches have significantly impacted its segment performance.
The Company does not believe that the value of unsold inventory has a significant impact on its segment results from operations because of the relative size compared to the size of the Company’s operations. At December 31, 2009 and 2008, the cost of finished inventory units were worth R$ 121.1 million and R$96.5 million, respectively, compared to a consolidated net operational revenue of R$ 3,022.3 million in 2009 and R$1,740.4 million in 2008, or a 4.0% and 5.5%, potential impact on the net operational revenue, assuming a hypothetical 100% sale of completed inventory units by the end of the period. The Company believes that the existence of such units is a natural part of its business and confirms that the focus of its business is to have as low inventory as possible.
Regarding changes in the mix, the Company does not believe that this has materially affected its historical results of operations. The company has minimum standards to approve each project and these standards do not change according to project mix. For example, in 2009, the company launched, under the Gafisa brand, a project called “The Place” in Goiânia - GO, with a total potential sales volume (PSV) of R$ 35.9 million for 17 units, or R$ 2.1 million per unit. In that same year, Gafisa launched a project called “Grand Park Árvores” in São Luiz – MA, with an average price per unit of R$ 183,900, or 91.3% below the project “The Place”. In the Company’s view, this is related to the characteristics of the regions where it is launching and not necessarily that the Company delivers lower margins on a lower average price per unit project. The lower 2009 gross margin, when compared to 2008, is related to higher than expected construction expenses, primarily in the north of the country and Rio de Janeiro, where we encountered lower productivity. This lower than expected productivity was mainly due to the lack of trained personnel, elevating the labor expenses, which already exceeded consolidated INCC (National Construction Cost Index), of 3.20% in 2009, compared to 7.74% of salary agreement.
Mr. Rufus Decker | 16 | March 4, 2011 |
10. | You proposed revisions to the Gafisa segment’s MD&A for the years ended December 31, 2009 and 2008 included a statement that the increase in operating costs was primarily attributable to the greater volume of construction in progress in 2009 as compared to 2008. However, your discussion of net operating revenue for the same period indicated that there was a decrease in the number of units launched from 4,949 in 2008 to 3,413 in 2009. It is unclear how the volume of construction for this segment increased during 2009 if the number of units launched decreased. Please revise to clarify. |
In response to the Staff’s comment, the revised disclosure should read as follows:
Gafisa
Results of Operations - Years ended December 31, 2009 and 2008
Net operating revenue
Net operating revenue for the Gafisa segment was R$ 1,757.2 million in 2009, compared to R$ 1,214.6 million in 2008, which represents an increase of 44.7% between the years. The increase in net operating revenues was attributable to the fact that Gafisa recognizes revenue under the percentage of completion basis, and the related increase in volume between years primarily is a result of an increase in sold units from inventory that was reduced in 2009. During 2009, due to the economic recovery of the financial crisis, Gafisa adopted a conservative policy which reduced the launch volume for the year and the number of units under construction with lower sales performance, consequently reducing the total volume of inventory, maintaining the company financial liquidity. At December 31, 2009, Gafisa had R$ 1,114.3 million in inventory and 14,775 units under construction compared to R$ 1,340.6 million and 18,864 units at December 31, 2008. This was demonstrated by an increase in “contracted sales” during the year. The Gafisa segment had contracted sales of R$ 1,510.1 million in 2009 compared to R$ 1,345.1 million in 2008.
The increase in contracted sales between the years reflects the continued strengthening of the Brazilian economy during 2009 and Gafisa’s ability to contract many of the developments that it had launched in 2008 and before. Despite this increase in contracted sales, there was a decrease in the number of units launched from 4,949 in 2008 to 3,413 units in 2009 which does not affect the net operating revenue, since Gafisa’s revenues are based on the percentage of completion for units sold and under development. This decrease reflects Gafisa’s ability to adjust its launches and inventories to optimally meet unit demand. The average price for a Gafisa unit launched during 2009 was R$ 370.5 thousand as compared to an average price per unit of R$ 386.6 thousand in 2008. The lower average price per unit in 2009 was attributable to a change in the project mix.
Audited Consolidated Financials
Note 25 – Supplemental Information – Summary of Principal Difference Between Brazilian GAAP and U.S.GAAP.
(c) U.S.GAAP Supplemental Information, page F-82
11. | We note your response to comment six from our letter dated November 23, 2010. It appears that during 2007, 2008 and 2009 you changed the structure of your internal organization in a manner that caused the composition of your reportable segments to change. Please tell us how you considered ASC 280-10-50-34 through 50-36 in deciding not to restate your segment information for prior periods so that the presentation between periods would be more comparable. |
In response to the Staff’s comment, the Company notes that it presented the following reportable segments during the following periods:
Gafisa segment (high and middle income units) | Alphaville segment (residential lots) | Bairro Novo segment (low income) | Tenda segment (low income) | FIT Residential segment (low income) | |
2009 | X | X | — | X | — |
2008 | X | X | X | X | X |
2007 | X | X | X | — | X |
As disclosed in both the notes to the Company’s 2007 consolidated financial statements and in its 2007 Form 20-F, in 2006 Gafisa reported its segments on a geographic basis. In 2007, it switched its segment reporting to an operational segment presentation in order to align with acquired businesses.
The Company does not believe that it significantly changed the way it conducted its business during 2009. It has and continues to focus its efforts on the sale of high and middle income units through its Gafisa segment, the sale of residential lots through its Alphaville segment, and the sale of low income units. Low income units have been sold through Bairro Novo, Tenda and FIT Residential. The Company has not consolidated those three entities into one low income segment for the periods presented because each of the entities had different group ownership percentages during the periods presented. Accordingly, the Company believed their separate presentation was more meaningful to financial statement users. The Company also notes that Bairro Novo and FIT Residential segment disclosure amounts were relatively insignificant during the periods presented. Accordingly, and with the incremental disclosures set forth in its February 14, 2011 response letter, the Company believes that it has complied with ASC 280.10.50 for the periods presented. |
(d) U.S.GAAP condensed consolidated financial information, page F-88
Mr. Rufus Decker | 18 | March 4, 2011 |
12. | We note your responses to comments seven and eight from our letter dated November 23, 2010. Please supplementally provide us with statements of cash flows in compliance with IAS 7 for the periods presented in your Form 20-F so that we can understand how these cash flow statements would have differed from the Brazilian GAAP cash flow statements presented in your filing. Please also provide us with your materiality analysis under SAB Topic 1:M explaining how you determined that the errors in your IAS 7 cash flow statement and the presentation of your cash and cash equivalents under U.S.GAAP were not qualitatively or quantitatively material. |
As set forth in response to comment 1 from the Staff’s letter dated February 18, 2011 , the Company intends to revise the US GAAP footnote of its historical financial statements to retrospectively account for certain financial instruments as marketable securities instead of as cash equivalents and to otherwise correct any material matters in response to the Staff’s comments. As part of the revision discussed in the response to comment 1, the Company will include a cash flow statement complying ASC 230.10. The Brazilian GAAP cash flow statement will be revised in connection with the implementation of new Brazilian GAAP accounting rules in 2010.
* * * * * *
As requested, Gafisa acknowledges the following:
· | Gafisa is responsible for the adequacy and accuracy of the disclosure in the filing; |
· | Gafisa comments or changes to disclosure in response to Staff comments do not foreclose the Commission from taking any action with respect to the filing; and |
· | Gafisa may not assert Staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States |
Gafisa would appreciate receiving any further questions or comments that the Staff may have regarding this letter at the Staff’s earliest convenience.
Sincerely yours, | ||
/s/ Alceu Duilio Calciolari | ||
Alceu Duilio Calciolari | ||
Chief Financial Officer and Investor Relations Officer | ||
Gafisa S.A. |
cc: Manuel Garciadiaz, Esq. (Davis Polk & Wardwell LLP)
Daniel Gomes Maranhão Junior (Ernst & Young Terco Auditores Indep. Soc. Simples)