Green Brick Partners (GRBK): A Cheap, Complicated Homebuilder Focused on Dallas and Atlanta
I chose to write-up Green Brick Partners (GRBK) this week for a couple reasons. The first is the company’s headquarters: Plano, Texas. I live in Plano. And the company gets about half of its value from its Dallas-Fort Worth homebuilding operations. My “initial interest post” checklist goes something like this:
- Do I understand the business?
- Is it safe?
- Is it good?
- Is it cheap?
The single most important questions is number zero: “Do I understand the business?” Since I’ve lived for about seven years right by this company’s lots – I should understand it better than most homebuilders. The other half of the company, however, is in the Atlanta area. That is a place I know nothing about. So, the answer to question zero would be that I understand half the business here well.
The next easiest question to answer – after “do I understand the business?” – would be #3 “is it cheap?”.
So, we’ll skip right to that one. It is, after all, the other reason that put Green Brick Partners at the top of my research pipeline.
I have in front of me the balance sheet for Green Brick Partners dated September 30th, 2018. This is the last day of the most recent quarter the company has provided results for. Under “inventory” we see $648 million. Under “cash” we have $33 million. There’s another $12 million under “restricted cash”. The unrestricted part of cash is offset almost exactly with customer deposits. The restricted part of cash is just $12 million. Debt is about $200 million gross. So, that leaves about $188 million in net debt. If we netted out that inventory less that net debt we’d be left with $648 million in inventory less $188 million in net debt equals $460 million. The company has a little less than 51 million shares outstanding. So, $460 million in real estate free from debt divided by 51 million shares outstanding equals $9.02 a share. Let’s call that $9 a share. That’s very close to the company’s officially stated net tangible book value of $8.97 a share. Again, that’s basically $9 a share. We can compare this to the market price of $8.06 a share at which GRBK stock closed today. So, we have a stock with tangible book value – almost all land (about 50% in Dallas Fort-Worth and about 50% in the Atlanta area) – of $9 a share against a market price of $8 a share. Green Brick Partners is trading at about 90% of book value. So, a price-to-tangible-book ratio of 0.9 looks cheap.
However, this is where we start getting into the more complex aspects of Green Brick Partners. The company’s balance sheet shows only $15 million (about 29 cents a share) in “noncontrolling” interests. Green Brick, however, has only a 50% economic interest in its Dallas Fort-Worth and Atlanta homebuilders. The fair market value of the 50% owned by its partners – basically, the top management of these “controlled builders” – would be far more than $15 million. We know this because on the income statement we can see that the 50% of these homebuilders that Green Brick doesn’t own will probably report after-tax earnings of between $10 and $12 million for fiscal 2018. If we put a 15 times P/E on that half of those builders Green Brick Partners doesn’t own – the “noncontrolling interests” would be carried on the balance sheet at more like $150 million to $180 million than $15 million. That’s a big difference (it’s over $3 per share). So, does Green Brick Partners common stock have a tangible book value of more like $9 a share or $6 a share? This makes a big difference when the stock trades at $8 a share.
To answer this question, we’ll need to talk about Green Brick Partners’ accounting. And, to do that, we’ll need to talk about the way the company is set up. Let’s start with the owners.
As of the company’s proxy statement for the 2018 annual meeting, Green Brick Partners had 3 important owners: 1) Greenlight Capital, an investment company controlled by David Einhorn (the company’s chairman) owned 48% of the company, 2) Third Point Funds (which are Dan Loeb controlled funds) owned 16% of the company, and 3) Jim Brickman, Green Brick Partners founder and CEO, owned 4% of the company. During 2018, Third Point sold its shares of Green Brick Partners in a public offering.
How did Third Point get those shares?
This brings me to the other reason I’m doing an initial interest post on Green Brick Partners this week. I run an “overlooked stocks” portfolio for the accounts Andrew and I manage for Focused Compounding Capital Management. To count as “overlooked” a stock must meet at least one of several criteria. One of the boxes it can check is a “reverse merger”. Another box it can check is a micro-cap. Green Brick Partners is not quite a micro-cap stock (the company has a market capitalization of just under $400 million; though insiders – including Einhorn’s Greenlight Capital – eat up about $200 million of that market cap). But, it is a reverse merger.
Green Brick Partners was formed out of a former publicly traded ethanol company. That company sold its ethanol operations and became a public shell. Green Brick Partners was then merged into that public shell. This is how both Third Point and Greenlight ended up getting involved with Green Brick. First, there was a loan – from Greenlight – to finance the purchase of the homebuilding business. And then Greenlight and Third Point bought shares in Green Brick Partners to raise the cash used to eliminate that debt. Later – in 2018 – Third Point sold its shares to the public. From the time Green Brick announced Third Point’s intention to sell its stock till today, the stock’s price declined from more than $12 a share to less than $8 a share.
So, is this an “overlooked stock”? The reverse merger was 3 plus years ago. The stock is not a micro-cap stock (it’s around $400 million) but it is a mico-float stock (the float is less than $200 million). Did the company have an IPO? The closest thing to an IPO was Third Point’s sale of about $100 million (originally) worth of stock to the public. That “IPO” (really, a secondary offering – but, in this case, the first time shares were being widely sold to the public all at once) may have helped depress the stock’s price. This is debateable. The initial announcement certainly did cause the stock price to plummet (it dropped like 25% in a single day). But, the longer term decline in the company’s stock price over the past 6 months or more may be due as much to a something like 10% decline in the overall stock market coupled with industry specific (the homebuilding industry, that is) concerns about rising interest rates. Overall, Green Brick has some of the features that may make a stock count as “overlooked” or just plain disliked. So, early 2019 might be a good time to check it out.
What, then, is Green Brick Partners really? What do you – as a buyer of GRBK shares – get for your money?
The value in Green Brick Partners comes from its controlling stake in captive homebuilders in the Dallas Fort-Worth and Atlanta markets. Green Brick has some other assets: a 100% owned homebuilder (in the start-up stages) in DFW, an 80% owned homebuilder in Florida, a non-controlled (and thus unconsolidated) 49% owned homebuilder in Colorado, and a mortgage business. However, none of these other holdings contribute meaningfully to the value of the company. A couple of these other assets may contribute meaningfully starting in 2019 or 2020. But, for now, we will discuss only one type of arrangement in one type of asset. We’re talking about the half-ownership / full control arrangements Green Brick has with homebuilders in Dallas Fort-Worth and Atlanta.
Here is how this arrangement works:
- Green Brick has a 50% economic interest in the homebuilder
- Green Brick has 51% of the voting rights in the homebuilder
- Green Brick appoints 2/3 of the board of the homebuilder
- Liabilities of the homebuilder are non-recourse to Green Brick
As I mentioned earlier, this complicates the accounting for Green Brick. It also changes the economics a bit too. Green Brick is more of a land developer and less of a homebuilder – though it’s important not to overstate this distinction – than you might think.
Why?
Green Brick – the corporate parent you are buying shares in when you purchase GRBK stock – has a 100% interest in the capital related activities of providing land and financing to controlled homebuilders. However, Green Brick only has a 50% economic interest in the actual homebuilding activities of these builders.
Technically, to make money: Green Brick can either transfer its land from the parent to a 50% owned homebuilding subsidiary or it can sell the land to an unrelated party (usually to a homebuilder in Dallas or Atlanta that it has no stake in). In practice, Green Brick almost always transfers the land to a 50% owned homebuilding subsidiary.
Most of Green Brick’s land inventory is not finished homes. If you look at a breakdown of Green Brick’s inventory it is about 50% held in the “land development” division and about 75% of the company’s inventory is classified as either “land held for development” or “work in process”. Only 25% of the inventory – by dollar volume – consists of finished lots held for sale to customers. Furthermore, we need to consider the distinction between the accounting presentation and the economic reality here. The inventory shown as lots held for sale are only really 50% owned by Green Brick. However, the land held for development starts out as 100% owned by Green Brick. For accounting purposes, Green Brick consolidates the accounts of its 50% owned subsidiaries. This means that if you see a 50/50 break down between land held by the “land development” division and land held by a homebuilding subsidiary – the land division assets are 100% owned by you (the GRBK shareholder) but the homebuilding land is only 50% owned by you. This land is, of course, ultimately going to be the exact same land. At an early stage it is 100% owned by you and then at a later stage – probably anywhere from 2-6 years down the road – it is only 50% owned by you (and 50% owned by the managers of the homebuilding subsidiary).
So, let’s look at Green Brick’s inventory again. As of last quarter, the company has about $650 million in land on its books. This land is carried at cost unless marked down for impairments. Green Brick hasn’t really marked anything down in the last 4 years or so (the markets Green Brick builds homes in have been rising in price for about 7 straight years or so). The cost shown is not just the price Green Brick paid for the land. Green Brick capitalizes its interest and adds this to the cost of the land it holds. Development costs are also added to the land. This can include all sorts of costs related to getting permission to develop the land and things like that. Nonetheless, you would expect that – at least as of today – the $650 million at which the land is held on Green Brick’s books is an understatement of the fair market value of the land.
Why?
Well, for one, Green Brick hasn’t written down land recently. If the value of the land declined below the cost on the books – Green Brick would have some impairments. Two, Green Brick consolidates its 50% owned homebuilders that eventually build on these lots and receive cash from customers. The homebuilding subsidiaries have recently had the following margins: 22-23% gross margins, 12-13% operating margins, and (adjusted for certain one-time tax law changes and use of net operating loss carryforwards) 9-10% after-tax margins. In other words, these homebuilders have recently been selling the land for about 30% more than their cost (0.3/1.3 = 23% gross margin). So, on a fully consolidated basis, Green Brick’s inventory is eventually monetized at about 1.3 times its book value.
The inventory is – again, we’re talking the consolidated accounting treatment here – split almost exactly evenly between land held by the land development division (which is 100% owned by GRBK shareholders) and land held by the homebuilding subsidiaries (which is only 50% owned by GRBK shareholders).
Let’s think of it this way. Green Brick shows $650 million in land inventory on its books. In reality, this land might be worth something like $850 million ($650 * 1.3 = $845 million). That means the company might consolidate about $650 million in land at cost and about $850 million in land at its likely fair market value. However, this accounting consolidation doesn’t match the economic reality. The economic reality – for a GRBK shareholder – is that you own 50% of $325 million of land at cost (and 50% of $425 million of land at its likely fair market value) and then 100% of the other $325 million of land at cost (and 100% of the $425 million of land at its likely fair market value).
So, $325 million times 0.5 equals $163 million. And $163 million plus $325 million equals $488 million. That’s at cost.
At our admittedly arbitrary valuation of about 1.3 times cost – the fair market value calculation would then be $325 million times 1.3 equals $423 million times 0.5 equals $211 million. And then $211 million plus $325 million times 1.3 ($423 million) equals $634 million.
That is, however, the gross value of the inventory. It doesn’t take into account the company’s debt. GRBK borrows about $200 million and has about $45 million in cash. Much of the cash is offset with customer deposits. However, it’s probably best to think of these customer deposits as being secured by homes soon to be delivered rather than by the company’s cash. This is because 80-85% of the orders for homes will not be cancelled. So, even if every order that is cancelled results in the customer getting back 100% of their deposit (which seems unlikely), over 80% of the deposits will be kept and a home delivered to satisfy the obligation caused by taking the deposit.
For this reason, I’m going to treat the company’s net financial debt as being just $155 million ($200 million in debt less $45 million in cash). I don’t see customer deposits as a financial liability. It’s really just a reduction in the amount of net assets that have to be tied up in homebuilding operations.
Okay. So, we have $155 million in net debt. And we – as shareholders of GRBK – have a “look through” interest in about $488 million in land at cost and about $634 million in fair market value of the land. We can subtract the net debt from the “look through” land. So, at cost, that would be $488 million less $155 million equals $333 million. We then divide $333 million by 51 million shares outstanding to get $6.53 a share in “book value”. I think – restated for the 100% ownership at the land development level and 50% ownership at the homebuilding level – this roughly $6.50 is an accurate enough estimate of the economic (as opposed to accounting) net tangible assets of GRBK. This stock really has a tangible book value of $6.50 a share.
That’s at cost. What’s the likely fair market value of these assets? That would be $634 million of land minus $155 million of net debt equals $479 million. Then we divide $479 million by 51 million shares outstanding. We get $9.39 a share. That’s a decent estimate of the fair market value of what a GRBK shareholder really owns.
The stock closed today at $8.06 a share. So, GRBK is probably trading at about 1.23 times tangible book value ($8.06/$6.53) and about 0.86 times appraisal value ($8.06/$9.39).
Is it possible that I have overstated or understated the fair market value – the $9.39 “appraisal value” – of the stock here by a meaningful amount?
Yeah. It’s possible. The issue here is really in the land development division. This is the 100% owned business. The homebuilding division is less of a concern for a few reasons. One, it only takes about 5 months to 9 months to build a house. Two, the land being sold has been transferred by GRBK from the land development division to the homebuilding subsidiary to generate a targeted rate of return at the parent company level. Three, we can see from an accounting breakdown of net income attributable to non-controlling interests that the homebuilding division creates less income than the land development division. For example, over the last 9 months, GRBK reported $52 million in operating profit and yet only $9 million in net income attributable to the non-controlling interests. This means the non-controlling interest account for about 20% of normalized after-tax profit (here, I’ve normalized the tax rate as if it was 21%). If the land development division and the homebuilding division produced exactly equal profits, the non-controlling interest should be 25% of normalized after-tax profit (because that’s 50% of 50%). Since the non-controlling interest is 20% of after-tax profits, this suggests that land development is more like 60% of GRBK’s profits and homebuilding is more like 40%. But, this could easily be the result of cyclicality. In an up market, GRBK may make more money from land development and in a down market it might make more from homebuilding. I just don’t know.
There’s also evidence from the returns that GRBK targets and some information I have about the rate of land price appreciation in Dallas and Atlanta and the length of time between when GRBK contracts to option land and when that land is finally developed enough to be transferred to the homebuilding division that suggests to me that the inventory in the land development division (which is 100% owned by GRBK) is carried at a lower value relative to its likely fair market value than the land that has already been transferred to the homebuilding division. Here, I used a 1.3 times multiple as if all of the land would – if sold today in whatever state it is in – generate the same gross margins as the 22-23% gross margins in the homebuilding division.
In other words, GRBK might deserve an appraisal value of more than $9.50 a share. But, I think it’s reasonable to use roughly $6.50 a share as “cost” and roughly $9.50 a share as “appraisal value” for this stock. The market value is about $8 a share on the NASDAQ right now.
Why haven’t I talked about earnings?
I just don’t think that’s a meaningful way to look at the business. One, the company does not pay dividends and has no plans to ever pay dividends. Two, it has not historically bought back stock and has only authorized a small (less than 10% of the company’s market cap) plan over several years to allow stock buybacks. Generally, all cash generated by this business will immediately be put back into buying more land. Therefore, “earnings” in GRBK are just the growth in per share land inventory. That’s it. You get paid in land in this company. Land will beget more land. Buying GRBK is an all-in bet on residential land in Dallas Fort-Worth and Atlanta where you will constantly let your bet ride on more and more land. You’ll never get cash from this stock. You’ll only get more and more land.
The other issue is debt. Homebuilders use debt. GRBK uses debt. It uses less debt than some other homebuilders. And, if the company was to quickly grow EPS, it would do it by increasing its debt/capital ratio from something like 20% to something like 40%. This boosts earnings, but it wouldn’t distort my calculation of land at cost and appraisal value. I think my approach is better, because it normalizes the cyclical distortions of leveraging and de-leveraging to fund growth faster than the sustainable rate of growth the company can support during boom years and to fund slower than the maximum sustainable rate of growth the company can support during bust years. Basically, I like my asset value approach better because it is less likely to overvalue this stock during a housing expansion and undervalue it during a housing contraction.
Finally, there is a tax issue at GRBK. This distorts earnings. And it’s easier than normal here for an analyst to be fooled by after-tax earnings. Income attributable to non-controlling interests appears below the tax line. Interest appears in the cost of goods sold line (because – as a homebuilder – GRBK capitalizes interests and assigns it to specific lots it sells). Two things have happened with taxes at GRBK. One, the company had – but, going forward, soon will not have – net operating loss carryforwards. Two, the corporate tax rate was cut from 35% to 21%. This tax cut caused a huge one-time tax distortion at GRBK (like it did at many public companies). GRBK also has a short history since the reverse merger. The reported EPS here just hasn’t been very meaningful. The company provides adjusted numbers like the EBITDA return on average equity.
A more useful figure is to look at things like the return on capital at GRBK and at other homebuilders. And then to look at the leverage normally used at GRBK and at other homebuilders. Overall – and I may be being ever so slightly conservative here – I think the likely long-term return on equity here is similar to the likely long-term return in the stock market. Specifically, I would say that if you use my appraisal value (of almost $9.50) which is basically equivalent to a sort of “mark-to-market” approach to the company’s assets – the compound annual growth rate in this company’s appraisal value per share, book value per share, etc. shouldn’t be much greater than the return in the S&P 500. It might be as fast as 10% a year. It’s not – over a full cycle – going to be something like 20% a year.
So, GRBK might be an adequate investment at a price equal to my “appraisal value” of $9.39 a share. I would want a margin of safety. So, a good price to buy the stock would be something like two-thirds of appraisal value. The formula for this – at present – would be $9.39 * 0.65 equals $6.10 a share. The cost approach gives you an adjusted book value of about $6.53 a share. This would be about 70% of my initial appraisal value per share. That would be another logical line in the sand. So, let’s use $6.50 – as of this past quarter – as a good price at which to buy GRBK.
I didn’t talk much about the kinds of returns GRBK generates. We only have data for the expansion years. Atlanta and Dallas have both been expanding – homes have been rising in price – for the last seven years. GRBK does have better margins, returns on capital, lower leverage, etc. than other homebuilders. However, GRBK (the stock) is about two-thirds economic profits from land development and one-third homebuilding. It’s only in the hot housing markets of Dallas and Atlanta. And we only have data from the recent housing expansion. I don’t want to assume the company’s quality is higher than the average homebuilder.
So, how would I answer the 4 questions that matter most in an initial interest post?
- Do I understand the business? – I understand the Texas part of the business fine (a lot of the company’s lots it will develop over the next 5 years or so are in locations very close to where I live – not just in DFW generally, but specifically in the towns and parts of towns I know well). So, 50% I understand well and 50% (Atlanta) I know nothing about.
- Is it safe? – Homebuilders are cyclical, sensitive to interest rates, and use debt. GRBK uses less debt than many homebuilders. We are very far from the past peaks in homebuilding. The time to be least concerned about a housing bubble is when people still remember the last one. The long-term economics of supply/demand etc. in the DFW market are good (I don’t know Atlanta). Homebuilders aren’t very cash generative. The company has very big theoretical off balance sheet obligations (basically, options on land) I didn’t discuss – but, it’d be easy to back out of these at a fraction of the value of the land they have options on. Overall, safety here is – at this point in the cycle – adequate but not excellent.
- Is it good? – GRBK is a homebuilder. That’s a cyclical industry that is capital intensive. It’s not an especially bad industry. But, it’s definitely not an especially good industry. So far, GRBK’s returns on capital are better than homebuilders generally. Land development returns as an asset class of sorts though aren’t really better than the stock market generally, especially when you adjust for the use of leverage. GRBK’s quality is adequate but not excellent. I’m not sure I’d call it a “good” business. But, I don’t have evidence it’s a “bad” business either. It’s okay.
- Is it cheap? – GRBK stock is cheap. The adjusted book value is about $6.50. My appraisal value is about $9.50. The stock trades at about $8. So, the stock trades at about 125% of book value and about 85% of appraisal value. Today, most stocks trade above my appraisal value of the stock. Certainly, most stocks trade at much higher ratios of book value than 1.25x. GRBK is cheap.
Overall: GRBK is a cheap stock that I understand somewhat well enough and is of average quality and safety. It’s a value stock. And it’s a value stock where half of the value is in locations I am very, very familiar with. But, that’s about it. I’d first consider buying GRBK at $6.50 a share – and start to get serious about the stock at $6 a share – but, I’m not interested at prices above that.
Geoff’s initial interest level: 60%