While the homebuilders’ stocks have struggled lately and largely stalled out over the past two years as numerous home construction indicators decelerated and stalled out, a strong case can continue to be made that housing should reaccelerate given how low volumes remain in a historical context and relative to normalized demand.  However, this write up is less about the homebuilders in general and more about Pulte in particular -- I think PHM is more of a special situation, and the most misunderstood of the larger builders.

PHM has undergone a very significant turnaround and business transformation over the past 3-4 years.  Gone are the days of focusing on size, unit volumes, and margins without regard for asset turnover and capital efficiency.  CEO Richard Dugas now sounds like more of a salesman for EVA than anything else – and meeting with him or quarterly conference call now inevitably comes replete with multiple comments about returns.  While repetitive, there is a lot to be said for infusing the organization with a clear message and marching orders!  Over the past 3-4 years, the focus on returns and the underlying components of margins and asset turns have been nothing but consistent. 

I had been highly skeptical and virtually unwilling to believe any transformation could have taken place at PHM given the company’s longer term history and the fact that the CEO presiding over the bubble and crash remains in place.  Yet the changes in the past 3 years are both undeniable and massive.  Moreover, besides the massive business strategy makeover that came out of an extensive deep dive from one of America’s top consulting firms, the new CFO appears to be an additional enforcer of the new strategy and much of the legacy board of directors has been replaced.

Specifically, gross margins have gone from essentially worst to first, especially when normalizing for the treatment of sales commissions.  Asset turns have improved dramatically with a lot of room left to go.  The massive SKU and complexity reduction initiatives have borne fruit, but significant runway still remains as CMPs (commonly managed floor planes reached 39% last quarter, on the way to 70%.  Land acquisition underwriting now considers the cost of capital and length of project – there is no longer a focus only on margins.  In addition, there is no longer just an absolute IRR hurdle, but now the IRR hurdle is adjusted for risks.  What that means is that a higher IRR is required to justify a more asset intensive project with slower turns.  This all seems like basic stuff and common sense, but nonetheless it represents a sea change.  Homebuilding debt has fallen already by ~$2.5 billion, while the net debt reduction has been larger still.  Yet the company has significant excess cash on the balance sheet and will continue to generate significant incremental FCF in coming years.

While the business transformation has largely already taken place, and while the financial transformation is underway, I believe that the financial transformation has a very long way to go from here.  In particular, as management continues to gain confidence in the strategy and as time continues to pass, I believe that the amount of capital to be returned to shareholders will grow disproportionately to a very, very significant level (to be detailed later).

It is no stretch to say that NVR has proven decisively over the past 20 years that it clearly has the best business model in the industry over the course of the cycle (asset light, high turns, utilize FCF for aggressive share shrink).  That said, it has been among the more challenged homebuilder businesses in recent years as would be expected in the earlier innings of the cyclical upturn when the relative supply of finished lots on offer from developers is low.  NVR’s geographic concentration has been a major boon over the past 20 years but also more of a headwind in recent years.  While it is far from certain and will need to be proven out over time, I believe that PHM could be the best of both worlds over this cycle, particularly if management acts more decisively as time continues to pass.  PHM today can still benefit from its long legacy land position at this point in the cycle when the relative supply of finished lots on offer from developers is low.  That helps PHM today when the NVR model is more challenged, as does PHM’s more diverse geographic footprint.  But over time, as the cycle progresses and land developers rebuild their access to capital and lot supplies, and as PHM continues to monetize and slim down its excess owned lots, PHM could shift its future mix of lots dramatically towards finished lot options and away from owned lots.  And PHM could be doing this at the same time that the relative supply of said lots on offer from developers should be increasing and thus providing better terms on a relative basis. 

Most builders tend to trade in tight bands anchored to book value since the business is so asset intensive and the assets suffer from rather rapid depletion.  PHM trades that way today, but at a slight discount (as shown below).  NVR trades at a big BV premium to peers due to its capital efficiency and dramatically higher turns.  But it has been that way for years and isn’t a surprise.  PHM is in the midst of a dramatic transformation that should shift its complexion closer to NVR over time – the full extent and speed will be heavily dependent on whether management is willing to act decisively going forward.  That said, while it is not a certainty, this provides a tremendous opportunity for both exit multiple expansion and continued capital return over time.  Importantly, this option appears to be free today.

PHM has a massive deferred tax asset today.  I believe this DTA is not adequately appreciated or understood because it is so unique among the builders. The size of the current DTA speaks volumes about the magnitude of capital destruction unleashed at the company during the prior cycle and under the prior business model.  It also speaks to the acute lack of urgency to immediately monetize its DTAs that PHM demonstrated several years ago when competitors like LEN and DHI were acting decisively to do so.  That said, the benefit to PHM shareholders from today is that this DTA remains ready to be harvested now in the next several years at PHM whereas it has essentially been harvested already at the other profitable builders (i.e. LEN, DHI, TOL, etc).  And for the handful of other builders who have massive DTAs but minimal profitability (i.e. HOV, KBH, etc), the PV of those DTAs is obviously lower as the timing and likelihood for realizing them is far more extended.

Unlike the other public homebuilders that are highly profitable today that are already full cash taxpayers or will be real soon, PHM will be the only highly profitable one that should not be a cash taxpayer for the next 4 years or so. The PV benefit of the DTA for the other public builders with high margins today is small in amount.  And stripping it out of those builders’ current stock prices for valuation purposes is questionable since shareholders of those builders won’t directly see it materialize as a capital return (as incremental investment in inventory $ should remain significant and offset it).  However, for PHM the significant DTA absolutely should materialize as a capital return in the next several years if management sticks to the new strategy (and indications thus far suggest they should do so).  Coupled with its coming shift towards lower asset intensity, there is no reason to believe that all the FCF derived from this cash tax benefit at PHM will be “lost” to incremental inventory $ investment.  Therefore, I believe it makes a lot of sense to remove it from the PHM stock price for purposes of calculating an implied PE multiple.

One of the concrete examples of where PHM will be reducing asset intensity and increasing asset turns in the coming years relates to its active adult division, Del Webb.  Mega-sized (2,000 - 5,000 lot) legacy Del Webb communities are a source of sunk capital that will be released in coming years as they are replaced with the newer version of smaller communities (i.e. 600 -1,200 lot).  Most other builders don’t have mega sized legacy communities.

One of the biggest potential sources of upside surprise is the possibility that PHM could deplete inventory $ on a net basis throughout the coming cycle to end up closer to NVR’s asset light model as I was alluding to earlier.  If so, that would release significant incremental capital beyond my modeled assumptions.  I’m still assuming in my model that inventory $ increase by $600-700mm each and every year going forward.  If that turns out to be conservative, the capital return potential could be much larger than I’m modeling.

In any event, we are likely to see very significant share repurchase activity in the coming years.  The logic is reasonably simply.  Management has been clear that they don’t want to raise the recurring dividend to a level where they ever would feel pressure to cut it in a downturn.  Already, PHM has the highest dividend yield other than MDC, which is more of an anomaly.  And I’m already assuming that PHM more than doubles the dividend by 2016.  They also seem rather predisposed against special dividends, despite my belief that it is a very sensible thing to do in a deeply cyclical industry at the top of the cycle.  Plus, I don’t think we are near the top of the cycle, and management has made it clear that they want to be “balanced” and not to try to make major timing decisions so that makes special dividends seem even less likely.  They seem pretty unlikely to embark on any transformative large M&A given their major strategy shift towards more of a widget manufacturing company and away from a decentralized land speculation company.  Any M&A is thus likely to be small and more akin to a supplemental land acquisition deal (similar to the recent acquisition of the Dominion communities).  Not to mention that very few targets would fit.  Finally, after the $2.5B+ of debt paydown that has already taken place, and with fully $1B of the remaining gross debt outstanding in the form of long dated investment grade unsecured bonds maturing in 2032-2035 with minimal covenants, there just isn’t much debt left to pay down even if they wanted to.  All that is left to pay down is $818mm of bonds maturing in 2015-2017.  I’m already assuming they do that, but they could easily refinance those as they come due.  So the option that remains for the large stream of FCF I am envisioning going forward after the process of elimination we have just gone through is, of course, share repurchase.  Moreover, net debt to capital is well below its peers and at 10% is far from management’s target of less than 40%.  And PHM recently put in place a $500mm undrawn revolver with an accordion feature that could take it up to $1B.

In the meantime, share buyback activity has picked up in the past 4 quarters but remains modest relative to the FCF generation.  Given management comments about not wanting to make a major timing call all at once, I would expect to see the buyback activity continue to ramp up over time.  Management has repeated on several occasions that they understand the large cash hoard is their largest non-earning asset today in the context of a company focused on returns.  Perhaps an activist gets involved if management drags their feet.  And this may be a stretch, but even an LBO firm with too much “cash on the sidelines” could potentially get involved if the business continues along the path towards a more asset light model with significant FCF?  After all, there is no special share class of shares with additional voting rights.  Stranger things have happened.

Here is one way to think about upside potential, using assumptions that leave room for further upside if management or an activist were to push for more decisive action than what I am modeling.

Catalysts

No hard catalysts other than continued execution and a steady ramp in share repurchase activity.

It seems unlikely to happen all at once in a given quarter, but if we were to see a big increase in buyback activity to a teens annualized level as a % of the float, perhaps that will focus people on the unique FCF potential at PHM.

Risks

Homebuilding cycle.  My belief is we are still a lot closer to the bottom than the top in terms of industry volumes and activity.  That said, if the US economy goes into recession, it may be difficult for the housing industry to avoid falling altogether.  However, I would expect such a housing downturn to be shallower and shorter lived than normal, and the following upturn to be more pronounced and longer than normal as the spring gets coiled.

Change in strategy.  Management loses discipline and decides to binge on land at the top of the cycle.  This seems unlikely given the CEO and CFO have demonstrated so much “religion” on the new strategy and been so unequivocal and vocal in their support for it and in their admiration for NVR.

Risk to the strategy itself.  Margins on an asset light strategy ought to be lower than margins on an asset turn strategy if markets are efficient.  In that case, PHM gross margins going forward could fall more than I’m modeling.  That said, I would note that I’m already modeling gross margins before interest expense to fall going forward.  I would also note that NVR margins in the last cycle were the second highest in the group, so it is hard to say that markets are always efficient and that the asset light strategy should always have lower margins.

No hard catalysts other than continued execution and a steady ramp in share repurchase activity.

It seems unlikely to happen all at once in a given quarter, but if we were to see a big increase in buyback activity to a teens annualized level as a % of the float, perhaps that will focus people on the unique FCF potential at PHM.