Gafisa (GFA) is one of Brazil's largest residential home builders, and the company has been surprising investors of late with its string of positive quarters and rising cash flow. Its preliminary 4Q2012 results on January 15, 2013 posted its third consecutive quarter of positive cash generation. The company's guidance is rising as well. It originally estimated 2012 cash at R$500ML to R$700ML. The actual number came in at R$980ML. (Conversion of R$ to USD = 0.51)
Gafisa Group previously announced two goals as part of its 2012-13 turnaround strategy: deleveraging the balance sheet and generating free cash flow from ongoing operations.
All this is good news for long-suffering holders of the shares. By the end of 1H2012, Gafisa had burned through cash for almost 2 years, the ADRs were trading at $2/share, and the company had lost 90% of its capitalization.
Gafisa's current state reminds me of PulteGroup (PHM) in 2011. Pulte had acquired a homebuilder in 2009 that concentrated on the entry-level market (Centex), and Pulte bled red for 2 years before turning it around. A similar thing has happened to Gafisa with its September, 2008 acquisition of Tenda. (More on that later)
But a builder's fortunes can turn on a dime, and it looks like that's happening here.
To evaluate Gafisa, I blended the Q3 & Q4 preliminary results wherever possible, calculated the figures in R$, and then did a currency conversion to $USD/share (see below). The trend in metrics was what I was looking for. Afterwards, I reviewed my analysis with Caiman Valores, a fellow Seeking Alpha contributor, and we decided to write two separate but complementary articles.
Although Caiman employed a discounted cash flow method (DCF) for his analysis, and mine was more a bottoms-up approach, I find it noteworthy that our independent results were only 5% apart: $10 to $10.22/share USD. As of this writing, the shares are trading at $4.13 USD.
The only other residential builder I know that is trading with this kind of discount to intrinsic value is Brookfield Residential Properties (BRP), a Canadian builder with a significant landbank in the San Francisco Bay Area, Coastal Ca, Denver and Washington, D.C.
On a technical basis, Gafisa looks very much like the SPDR S&P Homebuilder Index (XHB) in December, 2011, just before it broke out to the upside. Homebuilders respond en masse to homeowner demand when a real estate cycle turns upward. If the cycle is heating up again in Brazil in preparation for the World Cup and Olympics, it will lift all boats.
That's why Gafisa's final numbers from 4Q2012, plus a resolution of the Alphaville rumors and Tenda liabilities, could mark a turning point for GFA. The future of Brazilian interest rates also plays a factor here. Although as mentioned by Caiman Valores, mortgages still play a relatively small role in Brazil. (See below)
Regardless of approach, we are examining an undervalued company that is emerging from a deep trough. A leaner Gafisa enters 2013 ramping cash, growing in profitability, and actively resolving its legacy issues.
* * * * * * * * *
Let's take a look at the 3Q2012 results reported on November 13, 2012.
|Total Consolidated debt + Obligations||R$ 4.174 BL (pg. 17)|
|Consolidated Cash and Cash Availabilities||R$ 1.235 BL (ibid)|
Total receivables: R$ 8.566416 BL
|R$ 8.566416 BL (pg. 18)|
|Inventory at Market Value||R$ 3.03660 BL (ibid)|
|Total Assets (9/30/2012)||R$ 12.838016 BL|
|Total Liabilities||R$ (4.174 BL)|
|Net||R$ 8.664016 BL|
|Brazilian Shares||432.27 ML|
|Value Per Brazilian share||R$ 20.043|
|Value Per Share in U.S. dollars (0.51 x R$)||$ 10.22 USD|
|Debt Covenants (outstanding debt x 2.2)||R$ 9.1828BL|
|Total Assets (9/30/2012)||R$ 12.838016 BL|
|Level above covenants (Assets minus debt)||R$ 3.655 BL|
So what is Gafisa's story; and what should we look for going forward?
Gafisa is a vertically integrated builder. It has three divisions: Alphaville, Gafisa, and Tenda, each targeting a different level of buyer, from first-time buyer up to luxury custom homes. It is the only South American builder to trade on the NYSE.
In September, 2008, Gafisa Group bought a controlling stake in Tenda, a company which sells low-income housing to sub-prime buyers through government-sponsored banks. An analog in the U.S. would be Section 8 housing, except in Brazil this program assists first-time home buyers, not renters.
Brazil's low-income housing sector is an extremely difficult, but lucrative opportunity for a residential builder to navigate. Brazil has a large demographic of low-income families in urban settings. The government is committed to enhancing home ownership in this sector in an attempt to lift families out of poverty. And the societal difference between owning and buying a government-subsidized housing unit - rather than squatting or renting in transient settings - is dramatic.
But low-income housing brings with it its own set of difficulties. There's a high cancellation rate. Potential buyers can fall quickly out of pre-qualification. The bureaucratic over-reach in Brazil is astronomical (I read that a government-assisted mortgage in Brazil can be 280 pgs. long). Then there is the rampant uncertainty: government-funded projects often lead to cost over-runs because of delays. There's often a lack of skilled labor too (shoddy workmanship); and cheap building materials must be employed to cut costs. You can preview an ample selection of these Brazilian housing horror stories on You tube.
On the flip side, the government is desperate to do something about the problem and is throwing money at potential builders who will provide cost-effective housing for the working poor. The government also prefers to partner with large builders who can offer economies of scale (high-rise residential buildings) as a solution. The World Cup is coming in 2014 and the 2016 Olympics soon afterwards. Thousands of low income families are being displaced in Rio de Janeiro and Sao Paulo as these cities prepare for the events.
The potential market for low-income housing in Brazil is huge, estimated at R$ 70BL for 2013-14. Most of these units will be funded through a Federal government program called M.C.M.V. (My House, My Life) and disbursed through the bank Caixa Economica Federal. Caixa is a Brazilian analog to the U.S's Fannie Mae and Freddie Mac. C.E.F. loaned R$ 80BL in 2011 and R$ 100 BL in 2012. Half of all its lending is aimed at low-income housing
To read more about this, here is a May 17, 2011 article on M.C.M.V. from the Rio times "Minha Casa Minha Vida"
And this recent article from the Economist (Feb. 16, 2013), "If you Build it"
Gafisa's Tenda unit faces several difficulties. Its product must be cost-efficient to build; affordable to buy; and yet of high enough quality to foster customer satisfaction and keep community-wide construction defects to a minimum.
Tenda says it has had a positive experience with a new aluminum-mold construction technique it wants to deploy for its future projects. This process requires a smaller site workforce, and molders do not compete with conventional or structural masonry. The final product quality is less dependent on a skilled work force too (smaller quality deviations). Mold-construction yields a shorter job cycle (10 months vs. 15 months) and increases productivity per employee. Tenda notes that the Mexican low-income market has almost completely migrated from conventional masonry to molds during the last 10 years.
Here is a 6 minute video on the aluminum-mold construction method
Tenda's experience with Brazil's sub-prime market in 2010-2012 was disastrous, and it has yet to prove its new business model for this aluminum mold product. In reality, Tenda's potential for growth is dependent more on public policy than real estate conditions. It must keep future scale to a profitable level and maintain adequate controls over every aspect of its projects .
The company has said that future Tenda communities will only be launched in the presence of basic acceptable conditions: resolution of legacy issues, a suitable urban landbank, and adequate control over project financing - especially timelines for planning and construction.
Cost overruns had been a real problem for Tenda in the past. Because of this, Tenda will not be undertaking new launches until the remaining ones are wound down in 2013. The good news is that the deleveraging of Tenda has been rapid and its current inventory is now quite low. It will no longer be adding to the financial problems, and as inventory is sold off, generates cash.
Gafisa's Three Divisions
Below are some of the characteristics and financials of Gafisa's three divisions. You will notice that almost all building costs for Alphaville and Gafisa launches are prepaid by sales of units under construction before the project is even completed. Alphaville and Gafisa are doing extremely well. The company now reports that even Tenda is contributing to its positive cash flow.
But Alphaville is its leader, the star division, and the company thinks a strong indicator of the future prospects for the company in both quality and profitability. It is widely-diversified across Brazil and builds suburban style homes. It is for this reason that I am mystified by recent rumors it would sell such a profitable division. Management has repeatedly rebuffed any attempts at fire-sale pricing, affirming its belief that the company is significantly undervalued.
The Gafisa division plans to wind down its remaining legacy projects (many of these were low margin affairs in new markets), and will return its focus in 2013 to its traditional (and highly-profitable) high-rise market in Rio and Sao Paulo. No new Tenda projects will be launched until its current inventory is sold or disposed of. Much of what is affecting investor sentiment vis-a-vis Tenda is past history now.
Income segment: Middle and Upper Middle income
Unit Prices: R$ 100K to R$ 500K
Characteristics: Horizontal lot development, suburban setting, custom projects,
80% of $R launches are pre-paid by time of completion
Launches - Alphaville
|2007||R$ 237 ML|
|2008||R$ 313 ML|
|2009||R$ 420 ML|
|2010||R$ 741 ML|
|2011||R$ 972 ML|
|2012||R$ 1,343 ML|
Contracted Pre-Sales - Alphaville
|2007||R$ 238 ML|
|2008||R$ 300 ML|
|2009||R$ 377 ML|
|2010||R$ 599 ML|
|2011||R$ 842 ML|
|2012||R$ 1,108 ML|
Income segment: Middle and Upper Middle income
Unit Prices: up to R$ 250K
Characteristics: Vertical (high rise), Metro setting, custom projects
2010 - 2012: Over 90% of $R launches are prepaid by time of completion. Going forward, the Gafisa division is narrowing its geographical focus back to Sao Paulo and Rio de Janeiro (98% concentration)
Launches - Gafisa
|2007||R$ 1698 ML|
|2008||R$ 1913 ML|
|2009||R$ 1264 ML|
|2010||R$ 2155 ML|
|2011||R$ 2155 ML|
|2012||R$ 1609 ML|
Contracted Pre-sales - Gafisa
|2007||R$ 1329 ML|
|2008||R$ 1345 ML|
|2009||R$ 1510 ML|
|2010||R$ 1974 ML|
|2011||R$ 2180 ML|
|2012||R$ 1600 ML|
Income segment: Affordable Entry-Level
Unit Prices: R$ 80K - R$ 250K
Characteristics: Vertical (high rise), Metro setting & surroundings, Standardized products.
Launches - Tenda
|2011||R$ 398 ML|
|2012||R$ 0 ML|
Contracted Pre-sales - Tenda
|2011||R$ 330 ML|
|2012||R$ -74 ML|
Preliminary 4Q2012 results were announced on January 15, 2013. The full year report will be released this Monday, March 11th, followed by a conference call on Tuesday, March 12th to discuss last year's results. I expect the company will affirm the turnaround and raise guidance for 2013.
There will also be some announcement concerning Tenda's future launches. Tenda is at the crux of the problems the company has had with debt levels and a big contributor to cash burn. A resolution on the horizon there would be well-received by investors.
And Alphaville? A sale or IPO of the division keeps arising in the rumor mill. It is Gafisa's most profitable division and the only one that will remain diversified geographically across Brazil. Why would the company sell Alphaville at these low prices? If there is an announcement, it had better be good. Management did not take kindly to some unsolicited offers for company assets in Februrary, 2012. It was trading at $6.50/share at the time; 60% higher than today.
Because their goal is to delever the company, I think management will go ahead with the short-term debt repayment in 3Q2013 (R$1.3BL), but will also make an effort to roll remaining debts forward past 2015. It's cutting it close this early in a turnaround to pay out so much cash for a short-term debt, but I believe Gafisa knows it has the resources to generate sufficient cash during 2013 to replace it. It also sends a clear message to the bond rating agencies. Better debt ratios greatly improve its ability to borrow for longer term and at more reasonable terms.
Do I think investors should buy Gafisa now? Well, I own a boatload of shares and have been adding since $3.50. The company also received two upgrades in January. But my bullishness is tempered by the recent uncertainty in the shares. The stock dropped 25% on NO NEWS in February after some solidly positive news in January. What's up with that?
I think it is probably just skittishness before next week's annual report. And I can't "be" in Brazil - like I am here in California - to observe the housing market first hand. But if Gafisa's turnaround is real, then here (under $4.50) is a good entry point for risk-tolerant investors. That's 40% below my estimate for true book value. As Louise Yamada the great technical analyst used to opine, "By the time it looks good, you're already too late."