Hovnanian Enterprises Inc (HOVNP) 2016 Q3 法說會逐字稿

Good morning and thank you for joining us today for Hovnanian Enterprises' fiscal 2016 third-quarter earnings conference call. An archive of the webcast will be available after the completion of the call and will run for 12 months.

This conference is being recorded for rebroadcast and all participants are currently in a listen-only mode. Management will make some opening remarks about the third-quarter results and then open up the line for questions.

The Company will also be webcasting a slide presentation along with the opening comments from management. The slides are available on the investors page of the Company's website at www.khov.com. Those listeners who would like to follow along should log on to the website at this time.

Before we begin, I would like to turn the call over to Jeff O'Keefe, Vice President of Investor Relations. Jeff, please go ahead.

Thank you, Andrew, and thank you all for participating in this morning's call to review the results for our third quarter, which ended July 31, 2016.

All statements in this conference call that are not historical facts should be considered as forward-looking statements within the meaning of the Safe Harbor Provisions of the Private Securities Litigation Reform Act of 1995. Such statements involve known and unknown risks, uncertainties, and other factors that may cause actual results, performance, or achievements of the Company to be materially different from any future results, performance, or achievements expressed or implied by the forward-looking statements.

Such forward-looking statements include, but are not limited to, statements related to the Company's goals and expectations with respect to its financial results for the current or future financial periods, including total revenues, adjusted EBITDA, and adjusted income before income taxes. Although we believe that our plans, intentions, and expectations reflected in or suggested by such forward-looking statements are reasonable, we can give no assurance that such plans, intentions, or expectations will be achieved.

By their nature, forward-looking statements speak only as of the date they are made, are not guarantees of future performance or results, and are subject to risks, uncertainties, and assumptions that are difficult to predict or quantify. Therefore, actual results could differ materially and adversely from those forward-looking statements as a result of a variety of factors.

Such risks, uncertainties, and other factors include, but are not limited to, changes in general and local economic industry and business conditions and impacts of the sustained homebuilding downturn; adverse weather and other environmental conditions and natural disasters; levels of indebtedness and restrictions on the Company's operations and activities imposed by the agreements governing the Company's outstanding indebtedness; the Company's sources of liquidity, changes in credit ratings, changes in market conditions and seasonality of the Company's business; the availability and cost of suitable land and improved lots; shortages in and price fluctuations of raw materials and labor, regional and local economic factors, including dependency on certain sectors of the economy and employment levels affecting home prices and sales activities in the market where the Company builds homes; fluctuations in interest rates and the availability of mortgage financing; changes in tax laws affecting the after-tax cost of owning a home; operations through joint ventures and third parties; government regulations, including regulations concerning development of land, the homebuilding, sales, and customer financing processes, tax laws, and the environment; product liability litigation; warranty claims and claims made by mortgage investors; levels of competition; availability and terms of financing to the Company; successful identification and integration of acquisitions; significant influence of the Company's controlling stockholders; availability of net operating loss carryforwards; utility shortages and outages or rate fluctuations; geopolitical risks, terrorists acts, and other acts of war; increases in cancellations of agreements of sale; loss of key management personnel or failure to attract qualified personnel; information technology failures and data security breaches; legal claims brought against us and not resolved in our favor; and certain risks, uncertainties, and other factors described in detail on the Company's annual report on Form 10-K for the fiscal year ended October 31, 2015, and subsequent filings with the Securities and Exchange Commission.

Except as otherwise required by applicable security laws, we undertake no obligation to publicly update or revise any forward-looking statements whether as a result of new information, future events, changed circumstances, or any other reason.

Joining me today from the Company are Ara Hovnanian, Chairman, President and CEO; Larry Sorsby, Executive Vice President and CFO; Brad O'Connor, Vice President, Chief Accounting Officer, and Controller; and David Backsetter, Vice President Finance and Treasurer.

I'll now turn the call over to Ara. Ara, go ahead.

Thanks, Jeff. We are happy to report a pre-tax profit for the third quarter. And while we recognize it's small, it's definitely a step in the right direction.

I'm also going to be talking about our credit statistics, which have improved nicely during the third quarter, as evidenced by our EBITDA to interest incurred coverage increasing to 1.4 times. And we'll also talk about our recent bond transaction, which certainly improves liquidity and allows us to increase our land spend to grow our community count.

Let me start with the quarter. We made steady progress in most metrics. If you turn to slide 3 on the top left-hand corner, you can see that total revenues grew to $717 million, a 33% increase versus last year's third quarter.

In the upper right-hand portion of the slide, you can see that our gross margin was 16.9% in the third quarter of this year compared to 17.8% in last year's third quarter. However, our gross margin did increase sequentially from 16.1% during the second quarter of 2016.

In the lower left-hand quadrant, you can see that we continued to leverage our SG&A expenses during the third quarter this year. Our total SG&A ratio decreased 330 basis points to 9.3% during the third quarter from 12.6% in last year's third quarter. In the lower right-hand corner of the slide, we show that we had a $3 million income before income taxes and land-related charges compared to a $9 million loss last year.

Turning to slide 4, you can see that some of the improvements we have made in our credit statistics. In the upper left-hand corner, we show that our adjusted EBITDA increased approximately 75% during the third quarter to $56 million compared to $32 million in last year's third quarter. And in the upper right-hand corner, we show that our adjusted EBITDA to interest incurred also increased to 1.4 times compared to 0.8 times in the third quarter a year ago.

Moving to the lower left-hand quadrant, our adjusted EBITDA for the nine-month period more than doubled this year to $135 million. While in the lower right-hand portion of the side, we show that over nine months, adjusted EBITDA to interest incurred also more than doubled, from 0.5 times to 1.1 times this year.

We expect a solid fourth quarter, with income before income taxes, excluding land-related charges and gains or losses from extinguishment of debt, to be between $32 million and $42 million.

Next, let's look at our gross margin. The drag on our gross margin continues to be a combination of factors, including rising labor costs, higher incentives and concessions in order to spur sales pace, and the impact of higher cost land we purchased in 2013 and 2014 when we and the industry were experiencing higher home absorption rates at our communities, and construction costs that were much lower.

As we continue to work through the land that we purchased in 2013 and 2014, we expect to see our gross margin improve. By 2018, we expect to have delivered on virtually all the land that we purchased during the 2013 and 2014 period. And therefore, we expect the gross margin drag from this land to decline in 2017 and 2018. Finally, increasing labor costs continue to be a problem in many markets and are a headwind for better gross margins for us and the industry.

We are not the only home builder, obviously, feeling pressure on gross margin. On slide 5, we show 10 of our peers who reported June or quarter July end results plus our own.

10 of the 11 home builders shown reported year-over-year declines in gross margin. We suspect the same combination of factors I just discussed that are negatively impacting our margins are also having adverse effect on our peers' gross margins. Fortunately, the decline in our gross margin during the third quarter was more than offset by continued improvements in our SG&A ratio, which we show on slide 6.

On the left-hand portion of the slide, we show our annual total SG&A ratio from 2001 through 2015. We consider 10% the ratio that we achieved in 2001 through 2005 to be a normalized ratio.

On the right-hand portion of this slide, you can see that our third-quarter SG&A ratio declined 330 basis points to 9.3%. Our improved ratio is also the result of our efforts to control SG&A expenses, improved sales pace per community, and a 33% growth in our total revenues.

During the first 9 months of 2016, our total SG&A ratio was 10.2%. So even before the benefit of our fourth quarter, which will be our largest delivery quarter, we've reduced our SG&A ratio back to almost normalized levels.

Now I'm going to switch gears and talk about our community count and the current sales environment. If you turn to slide 7, you can see that our consolidated community count has decreased 16% during the third quarter compared to last year's third quarter, from 206 to 174. This decline was partially due to the sale of ten communities in Minneapolis and Raleigh as we exited those markets, and the conversion of four consolidated communities into joint ventures.

In addition, we opened 83 new communities during the trailing 12 months and closed out of 101 older communities. However, you can also see that our joint venture community count has grown.

The sale of two of our divisions makes for challenging year-over-year comparisons for community count as well as net contracts. We'll redeploy the capital from selling Minneapolis and Raleigh and the proceeds from our new refinancings into our more successful existing divisions. This will lead to positive comparisons once these new communities are up and operating.

On slide 8, we show that despite a 16% decrease in our community count, both the dollar amount of net contracts and the number of net contracts only declined 4% during the third quarter. The sale of Minneapolis and Raleigh along with the phasing out of Tampa and the Bay Area of California makes the year-over-year net contract comparisons a less meaningful barometer.

We believe a more appropriate barometer of our sales results is to look at our contracts per community, a key metric for efficiency. We show that our consolidated contracts per community increased 14% to 8.4 net contracts per community for the third quarter.

As we've said on our conference call, we shifted our focus from growth to reaping operating efficiencies and improving our bottom line. Improvements in net contracts per community allow us to receive the benefit of those extra deliveries from existing communities without having to add much, if any, additional SG&A costs. And that certainly goes a long way to making us more efficient.

Slide 10 shows our net contract results re-stacked as if we had a quarter end so that we could compare our results to nine of our public peers who also reported results for a June quarter end. It's not surprising that with a 15% decline in our community count, their net contracts also decreased, but only by 2%.

This doesn't compare well to our peers, but you have to look at contracts per community for the whole story. If you look at contracts per community, as we do on slide 11, we had the largest improvement among the nine peers reporting.

Just to be clear: we show our net contracts per community for the same periods and the same period of time as the previous slide. Our 16% improvement in sales pace per community was the highest among our peers.

Slide 12 shows our net contracts per community on a monthly basis, giving as much granularity and transparency as possible. The most recent month is in blue; the same month a year ago is in gray. For 9 of the past 12 months, net contracts per community show year-over-year improvements and two months were flat.

For the third quarter of 2016, our net contracts per community were up 14% compared to the same quarter last year. The month of August shows a slight decline year over year, but there were five weekends in August of 2015 compared to only four weekends in August of 2016. So we're not overly concerned with a slight dip, particularly given our very strong third quarter.

On slide 13, we show on the left-hand side the dollar amount of our net contract backlog, including unconsolidated joint ventures, increased 8% to $1.5 billion. Despite the decline in community count, our backlog is still high and puts us in a good position for the strong fourth quarter.

I'll now turn it over to Larry Sorsby, our Executive Vice President and Chief Financial Officer.

Thanks, Ara. On slide 14, let me start with an update on Houston. Despite almost two years of significantly lower oil prices, our Houston division continues to post very good results.

During the third quarter of 2016, we saw the absolute number of net contracts in Houston increase by 6% year over year and net contracts per community increase 15% year over year to 6.9, which was the highest level for third-quarter net contracts per community over the past four years.

There are three things that set our Houston operations apart for many of the builders we compete against in that market. Number one: with an average home price on homes delivered of $306,000, we focus on a lower average price point.

Number two: we do not build in any of the highly competitive master-planned communities. And number three: we have less exposure to communities in the Energy Corridor in Houston than our peers.

It's been almost two years since oil prices have dropped and our margins and profitability levels in Houston remain strong. Despite another solid profitable quarter for our Houston operations, we remain cautious about the impact of lower oil prices on the Houston economy. We continue to keep a close eye on the market and we'll take appropriate actions should circumstances change.

Of note, our Dallas, Texas, operation, which is far less dependent on oil industry than Houston, remains strong as well. Houston and Dallas remain two of our strongest markets.

Turning to slide 15, you will see our owned and optioned land positions broken out by our publicly reported market segments. Our investment in land option deposits was $63 million as of July 31, 2016. Additionally, we have another $13 million invested in predevelopment expenses.

On slide 16, we show that our total lot count, including unconsolidated joint ventures, declined by almost 3,000 lots from the end of the second quarter to the end of the third quarter, partly due to our exiting from Minneapolis, Minnesota, and Raleigh, North Carolina, markets. You can see the reductions and additions on this slide.

We delivered about 1,600 homes. We walked away from 1,600 lots. 600 lots were sold with the land portfolios in Minnesota and North Carolina, and we added 900 lots through new option contracts. As we redeploy the capital from exiting these two markets and the proceeds from our new debt transactions, we expect to see the lots added from new option contracts and new land acquisitions increase in the future.

Looking at all of our consolidated communities in the aggregate, including mothballed communities and $281 million of inventory not owned, we have an inventory book value of $1.5 billion, net of $460 million of impairments. Our inventory, excluding inventory not owned, declined $177 million from the end of the second quarter to the end of the third quarter.

Some of the decrease is caused by a high volume of our third-quarter deliveries. However, the primary drivers of the decline are about $65 million related to converting 4 consolidated communities into unconsolidated joint venture communities and about $50 million from land sales, primarily related to our exit from North Carolina and Minnesota.

We are focused on return on inventory, and an important component of that is how quickly we can turn our inventory. On slide 17, you can see that we have the second-highest inventory turns over the trailing 12 months as compared to our peers.

In addition to our historical focus on more heavily utilizing land options than the majority of our peers, more recently, our turns were also aided by utilizing land banking, which facilitates us purchasing lots on a just-in-time basis. Achieving a high inventory turnover will continue to be a focus for us going forward.

Another area of discussion for the quarter is related to our deferred tax asset valuation allowance. During the fourth quarter of fiscal 2014, we reversed $285 million of a deferred tax asset valuation allowance. We should reverse the remaining valuation allowance when we began generating sustained profitability levels higher than recent years.

At the end of the third quarter of fiscal 2016, our valuation allowance in the aggregate was $635 million. The remaining valuation allowance is a very significant asset not currently reflected on our balance sheet and we've taken numerous stops to protect it. We will not have to pay cash federal income taxes on approximately $2 billion of future pre-tax earnings.

On slide 18, we show that we ended the third quarter with a total shareholders deficit of $152 million. If you add back the remaining valuation allowance, as we've done on this slide, then our shareholders equity would be a positive $483 million.

If you look at this on a per-share basis, it's $3.28 per share, which means that at yesterday's closing stock price of $1.98 per share, our stock is trading at a 40% discount to our adjusted tangible book value per share. Over time, we believe that we can repair our balance sheet, and have no current intentions of issuing equity anytime soon.

As seen on slide 19, after spending $132 million on land and land development in the third quarter, paying off $320 million of bonds that matured between October 2015 and May 2016, we ended the third quarter with $188 million of liquidity, which is within our liquidity target of between $170 million and $245 million.

I'm now going to talk about the capital market transactions we closed yesterday. For over a year, we've made comments that our preference was to refinance upcoming debt maturities rather than use our internally generated cash to retire them.

However, to date, the high-yield markets have remained closed, not just to us, but to most Triple C issuers. And we paid off rather than refinanced $320 million of our debt as it matured between October 2015 and May 2016.

Earlier this year, we began talks with H/2 Capital Partners that led to a series of transactions that we announced on July 29. The easiest way to describe these transactions is that they provided us with the capital needed to refinance the $121 million of 8 5/8% debt that matures in January 2017.

The first part of the transaction involved tendering for the debt that was scheduled to mature in January 2017. While we did not receive the required 90% participation to successfully complete the tender offer, yesterday we announced that we are calling all of our January 2017 notes at the make-whole price of approximately 102, about a point higher than our tender offer.

In order to fund this call, yesterday we issued 2 separate notes totaling $150 million, both of which were purchased by H/2. The first was a new $75 million super priority term loan at a floating rate of LIBOR plus 700 basis points, which at the moment equates to about 7 3/4%, that comes due on August 1 2019. The other one is $75 million of 10% second lien notes that come due October 15, 2018.

The collateral for both of these notes is the same as the collateral for the 7 1/4% first lien notes and 9 1/8% second lien notes due in 2020. We will be filing the indentures for these two notes as exhibits to our 10-Q this afternoon, but I wanted to make a brief comment on land banking restrictions that are being put in place as long as these two notes are outstanding.

There will be a $10 million quarterly limitation on land banking properties that are owned by us. However, there will be absolutely no restrictions on our ability to land bank newly identified parcels of land we do not own.

Additionally, as a final part of the series of transactions with H/2, we exchanged $75 million of our 9 1/8% second lien notes due November 2020 held by H/2 for a new 9.5% first lien note that comes due on November 15, 2020. These new first lien notes will be collateralized by the same assets that currently are held as collateral for our 2% and 5% notes due in 2021.

On slide 20, we show our pro forma maturity ladder, which takes into effect these new transactions. This series of transactions will allow us to increase our land spend over the near term rather than accumulate our own cash to pay off our $121 million of January 2017 maturities.

Over the prior 10 months, we have reduced our land spend due to paying off $320 million of debt. While the additional liquidity from the transactions we closed yesterday will provide capital to purchase land to increase our expected 2000 (sic) deliveries, the larger benefit will be to deliveries in 2018 and beyond.

We expect to shrink our consolidated deliveries in 2017 due to a combination of issues. First, our decision to exit from Minnesota, North Carolina, Tampa, and the Bay Area of California markets. Second, our decision to convert four wholly owned communities into JV communities and placing four additional communities we control by option that we initially planned as wholly owned communities into joint ventures, both of which we discussed during last year's quarterly analyst call.

Lastly, our less aggressive land spend over the prior few quarters also reduced our expected 2017 deliveries. However, when you look at our total deliveries expected in fiscal 2018, including both wholly owned deliveries and deliveries from unconsolidated joint ventures, we now expect 2018 total deliveries to be similar to what we anticipate to achieve during fiscal 2016. As we look forward, we need to improve our profitability levels, which will help improve our credit statistics and regain further access to the capital markets.

Turning to slide 21. We are refining our guidance for 2016. Assuming no changes in current market conditions, we expect to report total revenues of between $2.7 billion and $2.9 billion for all of fiscal 2016. We expect our gross margin for all of fiscal 2016 to be between 16% and 17%.

Additionally, we expect SG&A as a percent of total revenues for all of fiscal 2016 to be between 9% and 10%. We expect adjusted EBITDA to be between $200 million and $225 million for the year.

Excluding any land-related charges, gains or losses from extinguishing the debt, and other nonrecurring items such as legal settlements, we expect the pre-tax profit for all of fiscal 2016 to be between $25 million and $35 million. We feel like we're in a position to once again deliver solid performance metrics, and we look forward to delivering improved results for the remaining quarter of this year and beyond.

I'll now turn it back to Ara for some brief closing comments.

Thanks, Larry. As I said at the onset, we are happy to report a pre-tax profit the third quarter. And while the pre-tax profit is obviously small, it is definitely a step in the right direction. And we are very focused on improving our profitability levels from here.

As I also said, we're pleased that our credit statistics have improved quite a bit during the third quarter. And as I mentioned, a key measure for us, the EBITDA to interest coverage increased to 1.4 times.

Closing the transaction that Larry described with H/2 allows us to reinvest in more land sooner than we previously thought we'd be able to. The fact that we don't have to accumulate cash to pay for $121 million of debt that was due to mature in January of 2017 gives us the ability to redeploy that capital and invest in new communities. Ultimately, these investments will reverse a declining community count trends that we've seen in the past few quarters.

Our consolidated deliveries will shrink a little bit in 2017. However, we'll be reinvesting in land, as I said, sooner than we previously thought. And our 2018 total deliveries should get back to the 2016 levels. We remain focused on executing our business plans by closing out the year with a strong fourth quarter and beyond that, getting to higher levels of profitability over the next several years.

That concludes our formal remarks, and I'd be happy to open up the floor for questions.

(Operator Instructions) Sam McGovern, Credit Suisse.

Thanks for taking my questions. Just with regard to the financing transactions that you closed yesterday, I was hoping you guys could discuss your thought process there just in terms of the limitations in terms of using the baskets, et cetera, versus the way you previously funded things through land banking? How did you think about this transaction versus that?

We looked at this transaction very favorably. It's favorable rates as compared to some of our other ways of pulling liquidity levers. And I think it's a first step towards reaccessing the capital markets, which I think is also very important to the Company as a whole. And the limitation on land banking from our perspective is de minimis.

Historically, we typically would land bank parcels that were under our control but not yet owned. We really didn't do much land banking of owned properties.

However, last fall when the capital markets closed to us relatively suddenly and we had near-term maturities, we took a portfolio of land that we owned and land banked it. So we would kind of look at that as a one-off proposition for the most part, anyway.

So we don't think that limitation of land banking is a material limitation to us because going forward, as we identify a new parcels that we want to put under control if we decide to land bank it, there's nothing to prevent us from doing so due to these transactions.

Got it. And you highlighted that it does not restrict the land banking for some newly identified lots earlier in your prepared remarks as well. Should we expect new deals for that to be forthcoming, or are you just highlighting it just so people are aware?

I think that we'll make decisions as we look at individual parcels as to whether we want to do it wholly owned or whether we want to do it under a land bank. We now have cash that we weren't anticipating having prior to the H/2 transactions that we want to deploy first.

Got it. And for that $40 million of restricted cash that's identified for additional debt retirement, can you talk a little bit about where you guys would target? Would it be the near-term maturities at the end of 2017 or would it be for some of the notes that are trading at a lower dollar price?

Yes, I don't think I want to be any more specific than it's available to retire debt.

Okay, great. And then just as my last question, can you talk a little bit about target leverage. You know, how you guys get there and just where we should --

That's always an interesting point. I mean, one of the things that we learned in the last downturn is that our target leverage at that point was 50% debt to capital. We actually entered the home building great recession below that, and it wasn't low enough.

So it's kind of an interesting point because we are leveraged, like I like to say, to infinity and beyond since we have negative equity. So rather than set a new target right this second, we need to get back to our old target of 50%.

But frankly, I think that once we achieve that, it's more likely for us to set a target with a 3 handle, a 30% handle kind of targeted leverage. But we've got a ways to go before that's something that we can do on a near-term basis.

Great, thanks so much. I'll pass it along.

Alan Ratner, Zelman & Associates.

Thanks for taking my questions. The first one, just on the joint ventures. As the number of JV communities is increasing, and next year it sounds like you expect deliveries to increase quite a bit there as well, how should we think about the profitability coming from JVs? Because you have been printing losses there and those losses have been expanding as you're opening more JV communities.

Should we be extrapolating that out or is there a reason why that loss has been increasing over the past couple quarters? And how do you think about the overall profitability there?

Sure. This is Ara. The reason the losses you're seeing right now is we're starting many brand-new joint ventures. And they are at the start-up phase. We've got expenses associated with them, but not the deliveries yet.

The new properties that we are joint venturing we absolutely expect to be very profitable. And the structure of the joint venture is quite favorable for us while as long as we perform as we anticipate while our partner puts up the majority of the capital, we end up with a majority of the profit. So it's a good win-win partnership with GTIS, who we've done numerous ventures with and it's been a good relationship.

That's helpful. Thanks for that, Ara. And then second, just on the liquidity and kind of the conversation around cash. First, I know you have some outstanding borrowings on the revolver that's coming due in 2018. So curious if you've had any conversations with the lenders there or any plans for that.

And then just thinking about the proceeds you got from the H/2 transactions, you mentioned several times looking to deploy that cash. But really, with maturities now just kind of being pushed out into 2018 and 2019, it would seem like you would need to get that cash out pretty quickly and really have it be primarily finished lots that you could turn over the next, call it, 12 to 18 months in order to have that cash back to get ready for those maturities. Unless I'm thinking about that wrong.

So maybe just talk a little bit about that 2018, 2019 period. How quickly you think you could get that cash out and back in the door, and the revolver. Thank you.

Let me take the second question first, because that's the one I can remember. And then you may have to repeat the first question.

But we believe that as our performance continues to improve that we'll be able to access the capital markets again. So we are not contemplating that we're going to be shut down from refinancing the debt forever.

And I think we're optimistic that sometime between now and 2019, we'll be able to re-access the capital markets. Simultaneously, as our performance improves and we start to delever, maybe we don't refinance everything. So we are not thinking about this in terms of having to get the capital that we are redeploying back in a certain period of time.

Having said that, much of the new communities that we've been doing in recent years has been averaging 50, 60 lots per community and have averaged roughly a two-year life. So that's typically what occurs anyway, but not necessarily because we're looking to absolutely have that money back by the date that you were suggesting.

And I've been reminded of your first question really is the revolver. And on the revolver side of the equation, you know, we are not anticipating that we will be able to extend that revolver. We are anticipating paying that off in the summer of 2018.

Great, thanks for all of that detail. Good luck.

Nishu Sood, Deutsche Bank.

Hello. This is actually Tim Daley on for Nishu. Thanks for taking my question. My first question is in regards to the JV versus wholly owned communities.

Just trying to gauge when you say that the -- sorry, the full-year 2018 closings are going to be similar on a total level until 2016, could you break out how much of that will be coming from JV versus wholly owned communities? And compare that to what basically the breakout is for 2016.

No, we're not going to give you detail on that. What we're really trying to do is just tell you that we have reconfigured; we're going to be doing roughly the same volume in 2018 that we've done -- that we anticipate doing in 2016. And keep in mind that 2016 is a huge increase compared to what we actually achieved in 2015.

So as Ara and I like to say: we had two years' worth of growth in one year this year. Maybe it was three years' worth of growth in one year. And although we are going to take a little bit of a step backwards in 2017, by 2018, the combination of JV deliveries and wholly owned deliveries will be back to roughly the same level that we anticipate achieving in 2016.

But they'll be a little more weighted towards JVs and less weighted to wholly owned is about the only insight I'm able to give you at this time.

All right, that's very helpful. And then my second question. In regards to selling commissions, I was just curious as to could you put some numbers around how the commissions have trended over the past year? And then as well, due all your commissions feed to the SG&A line or is there any other allocation somewhere else?

To answer the second part first, all the commissions are in costs of sales for us, both internal and external commissions. And commissions as a percentage of revenue have not changed dramatically in the last year or so. It's been about the same.

The only thing I might add, we have seen incidences of the outside brokerage commissions getting a little additional incentive in some of the more competitive markets. But that's more anecdotal than a widespread trend.

All right, great. Thank you for that.

(Operator Instructions) Susan Berliner, JPMorgan.

Larry, I guess I wanted to start first with you with regards to the H/2 transaction. And it was helpful that you gave the land banking restrictions. Can you talk about any other restrictions, such as joint ventures or any holding onto X amount of inventory?

No other restrictions.

Okay. And the additional, you got some additional liquidity in the transaction. Can you say on a pro forma basis, have you used that to pay down the revolver?

Can I say if I have?

Yes.

I have not.

Okay. And then I guess, Ara, you had given some commentary with regards to forward guidance. And I wanted to focus on the margin because I know you guys aren't the only ones talking about labor issues.

But I guess as we look forward, are you expecting any improvement in gross margin going forward?

Well, as I indicated, we are still feeling a headwind from all the areas I mentioned. We did make a little sequential improvement and we expect to get some continued incremental improvement next year.

As I mentioned, part of that drag is from purchases we made in 2013 and 2014, much like many of our peers. Construction costs were much lower. They've gone up quite a bit since then, and paces have changed, although that's improving. But we think we'll expect our margins to gradually creep back.

I mean, it's interesting that it was just two years ago in 2013 and 2014 when our gross margin was 20%, which was a more normalized level by going back in our longer-term history. It has certainly eroded for many, including ourselves, but we are anxious to get it back to the levels that we just recently achieved.

And I guess following up on the SG&A. Obviously, you took it down a little bit for this year. Considering the makeup of joint ventures versus wholly owned, how should we think about the wholly owned SG&A margin in 2017 and 2018?

I would say that if you are able, and you're not able to do this, Sue. So I'll first say that if we treated the JV deliveries in 2017 and 2018 as if they were wholly owned and treated the management fee -- if we just looked at it as if it wasn't a JV and it was a consolidated community, our SG&A numbers are going to be similar to what we report in 2016. The fact that we have more JV communities delivering in 2017 and 2018, SG&A might tweak up a bit on a consolidated basis.

Overall, what's important for us is velocity per community, and that has a very big impact. That has been improving. So that's directionally the right way.

The other thing is as we are hoping to do a similar volume overall with four fewer divisions, that helps our SG&A as well. It's a little more efficient to have fewer larger divisions.

Great. And then I guess my last question would be just on the August orders. I know you discussed the difference in weekends. Can you talk about I guess the traffic in the communities? And I guess highlight any markets that did better than you expected or did worse? I know you talked about Houston and Dallas.

Overall, I'd say it's been relatively consistent across geographies. I don't think we can say we've had any real standouts.

Perhaps Maryland has been just a touch slower than some of the others. And we're a little bit behind in getting some new openings in southeastern Florida that we had counted on. The models are just running a touch behind. So that is part of it as well.

But all in all, we came off of a very strong third quarter, and we're not seeing any substantial shift in the marketplace. August is always a tricky period, anyway, given all the back-to-school activities, vacations, et cetera.

Great. Thanks so much.

I'll give you one more factoid. When you combine July and August together, sales per community were roughly flat year over year.

Any other questions?

No, that's great.

Thank you. No other questioners are in the queue at this time. So I'd like to turn the call back over to Ara Hovnanian for a closing remark.

Great. Well, thank you very much. We're making progress a step at a time, and we'll look forward to reporting a great fourth quarter as our year concludes. Thank you.

Ladies and gentlemen, thank you again for your participation in today's conference call. This now concludes the program and you may all disconnect at this time. Everyone have a great day.