Sunrun Inc (RUN) 2019 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q4 2019 Sunrun Inc.

Earnings Conference Call.

(Operator Instructions) As a reminder, this conference call is being recorded.

I would now like to turn the conference over to your host, Patrick Jobin, Investor Relations.

Please go ahead.

Thank you, operator, and thank you to those on the call for joining us today.

Before we begin, please note that certain remarks we will make on this conference call constitute forward-looking statements.

Although we believe these statements reflect our best judgment based on factors currently known to us, actual results may differ materially and adversely.

Please refer to the company's filings with the SEC for a more inclusive discussion of risks and other factors that may cause our actual results to differ from projections made in any forward-looking statements.

Please also note that these statements are being made as of today, and we disclaim any obligation to update or revise them.

On the call today are Lynn Jurich, Sunrun's Co-Founder and CEO; Bob Komin, Sunrun's CFO; and Ed Fenster, Sunrun's Co-Founder and Executive Chairman.

The presentation today will use slides, which are available on our website at investors.sunrun.com.

And now let me turn the call to Lynn.

Thanks, Patrick.

We are pleased to share Sunrun's fourth quarter and full year results and progress against our strategic priorities.

In 2019, we generated $102 million in cash, exceeding our annual target.

We also grew our customer base by 22%, while increasing adoption of Brightbox, our solar and battery service.

We added as many customers as the next 2 largest residential providers combined.

In the fourth quarter, we added 15,600 customers, representing 117 megawatts of deployments, a 9% sequential increase from the third quarter and the highest quarterly volume in the company's history.

At a 6% discount rate, we generated $100 million of net present value and created NPV per watt of $1.13 or over $8,700 per customer.

I'm pleased to report that we have worked through most of our construction labor bottlenecks, which limited our growth last year.

We've reduced the number of open positions in our installation organization from around 300 last quarter to about 100 today, in line with our growing business.

Open positions in our sales organization are now over 300, and this is typical as we ramp hiring into the busy summer season.

As the leading national company, Sunrun offers purposeful and mission-driven careers and a competitive total benefits package that allows us to stand out when we recruit.

We have aspirations to lead the development of the decentralized clean energy industry and are building differentiation through customer reach, customer experience and product differentiation.

For example, an existing big box retail partner has noted our differentiated execution capabilities, and we have the opportunity to expand into another 200 stores, which could grow our footprint with this retailer by over 20%.

We are also investing in battery attachment rates and grid services business development.

With many of these grid services programs, utilities will market to their customers on our behalf, and our customers will have access to new sources of value by sharing their batteries with the grid.

We have announced 5 grid services programs and have a strong pipeline.

Attachment rates for Brightbox sales in the Bay Area were over 50% in Q4, a level that has persisted in response to poor utility reliability.

In California, they were over 35%, and across all geographies, attachment rates approached 20% in our direct business.

Nationally, we have now installed over 9,000 Brightbox systems.

We expect Brightbox deployments to nearly double in 2020 compared to last year.

As always, we are focused on building industry's most valuable and satisfied customer base.

We maintain discipline on unit-level economics and deliver long-term value to our customers.

This is why we've achieved our market-leading position, and we intend to keep it.

We don't compete with dealer-only businesses who lack the capability to ensure a positive customer experience and who pay unsustainable prices.

We are exercising caution around the industry's recent acceleration of the direct selling or door-to-door sales channels through independent sales dealers.

I want to be very clear that this is an important acquisition channel, but it needs to have controls in management to prohibit aggressive practices that won't serve customers or investors over the long run.

The customer acquisition channels, including retail that we serve with our salespeople are growing over 20% and will be durable and provide cost reductions over time.

In sum, we believe we have continued to build the moat around the business, have worked through our labor issues and expect to see stronger growth rates in 2020 than we did in 2019.

We continue to expect long-term industry growth rates around 15% and expect we will grow at this level in 2020.

There is a groundswell building to find solutions to address the climate crisis.

Sunrun is enabling this transition today while providing households a superior energy service.

There is significant interest from corporates looking to engage with Sunrun as the category leader to invest with us and to partner with us to accelerate the transition to a decentralized energy system.

Not only a Sunrun committed to environmental, social and governance issues, which is core to our company philosophy and mission, but ESG awareness from capital and strategic partners is building and will continue to grow.

We are already a deeply carbon-negative company and seek to help our customers and partners become carbon negative as well.

We have also focused on many operational initiatives to deliver best-in-class customer value and to lower soft costs.

We launched our program called the Sunrun Way of Working to kick off the next phase of our operational improvements.

As part of these ongoing efforts, we are taking steps to reduce our installation cycle times, the time it takes from a customer signature through to install.

Beyond the obvious customer benefit, we believe fast cycle times are the key lever for driving down soft costs.

International markets in Europe and Asia have shown that with short cycle times, soft costs are as much as $7,000 less per customer.

For context, this is more than the scheduled ITC step-down.

Our initial efforts were focused on improving installation efficiency.

We completed time studies while also soliciting installer input, and the results have been significant.

In Q4, compared to the same period last year, we improved the construction labor efficiency by over 10%, more than offsetting wage pressures while maintaining our high-quality standards and commitment to safety and despite increased battery attachment.

With these installation-level improvements now standardized, we have extended our focus to include the full funnel from customer signature through to scheduling the installation.

These efforts include increasing technology investment, optimizing processes that we can control directly, for instance, site inspections, project management and install scheduling as well as externalities such as permitting.

I'd like to highlight 2 examples of these changes.

First, we have now implemented the use of drones for our site inspections.

The use of drones helped cut the total time by up to 50%, while reducing errors requiring revisits and improves the work experience of our site techs.

This also adds a wow factor for customers and their neighbors.

Second, in many jurisdictions, pipelines are affected by the local permitting process, which can create considerable waste.

One way we are tackling this waste is by driving permitting reform.

There is no reason permits can automatically be issued if they comply with industry standards as they are in leading international markets.

The solar automated permitting process campaign called SolarAPP will create a low-cost, seamless process.

Last year, the Department of Energy provided funding to multiple organizations to fast track permitting and interconnection, including funding to NREL to build an online permitting portal and partner with leading technical building code organizations.

The online portal will be piloted in 10 locations by this summer.

I am optimistic that over the next year or 2, we will have improved the process in most markets.

Las Vegas is an early adopter that instituted instant permitting and interconnection last year, and now the step of the process has been reduced from 30 days in 2018 to 0 in 2019.

With the benefit of this change, in early January, we installed a 27-panel system on a home in just 5 days after the customer signed up.

Records are meant to be broken.

And by the end of January, we were able to delight a customer by completing the installation just 2 days after sign up.

These initiatives, along with a continued focus on operational excellence, are laying the foundation for substantial cost reductions in the years to come, along with differentiated and improved customer value.

Turning now to 2020.

I am excited about our opportunities to further pull away from the pack.

We expect about 20% of growth in our customer base and 15% growth in new solar megawatts deployed.

Also, if you counted battery capacity the same way we count solar capacity, our growth rate would be about 25% in 2020.

We expect the combination of cash flow generation and net earning assets to grow faster than megawatt deployment growth.

For instance, about $100 million of cash generation and $190 million in net earning assets.

Due to the election, other global events and our strong balance sheet, this year, we may be selective in market timing for project finance transactions.

It's possible, we exit the year with more assets we hadn't turned out into the capital markets.

And if this is the case, we would see less cash generation but more net earning assets.

I'll now turn the call over to Bob Komin to review Q4 performance and to discuss guidance in more detail.

Thanks, Lynn.

NPV in the fourth quarter was approximately $8,700 per customer or $1.13 per watt.

For the full year 2019, NPV per watt was $1.05.

Project value was approximately $30,700 per customer or $4 per watt in Q4.

Now turning to creation costs on Slide 8. In Q4, total creation costs were approximately $22,100 per customer or $2.87 per watt, an improvement of $0.30 or 10% from the fourth quarter of 2018.

As with project value, creation costs can fluctuate quarter-to-quarter.

As a reminder, our cost stack is not directly comparable to those peers because of our channel partner business.

Blended installation cost per watt, which includes the cost of solar projects deployed by our channel partners as well as installation costs incurred for Sunrun built systems, was $2.25 per watt or $0.23 -- a $0.23 improvement from the fourth quarter of 2018.

Install costs for systems built by Sunrun were $1.96 per watt flat year-over-year.

In Q4, our sales and marketing costs were $0.69 per watt, down $0.11 from Q3.

Our total sales and marketing unit costs are calculated by dividing costs in the period by total megawatts deployed.

A higher mix of direct business results in higher reported sales and marketing costs per watt, but it also means there will be lower blended installation cost per watt over time due to the higher mix of Sunrun built systems at a lower cost per watt.

In Q4, G&A costs were $0.23 per watt, an improvement of $0.02 from Q3.

Finally, when we calculate creation costs, we subtract the GAAP gross margin contribution realized from our platform services.

This includes our distribution, racking and lead generation businesses as well as solar systems we sell for cash with a third-party loan.

Our platform services gross margin was $0.31 per watt in Q4, $0.05 higher than last quarter.

Our cash and third-party loan mix was 24% in Q4, slightly above recent levels, driven by increased demand ahead of the initial ITC step-down, which affected those systems that are customer-owned.

Leased systems are able to benefit from safe harboring to extend the ITC at higher levels.

We expect lease deployments to return to above 80% of the total mix in Q1.

The Q4 cost stack benefited from some seasonal and staffing dynamics.

In addition, some geo and product mix effects resulted in both lower project value and lower costs.

For 2020, we expect Sunrun built installed costs to improve modestly compared to the full year 2019, despite doubling the number of batteries installed.

We expect this benefit will be roughly offset by paying higher market rates to select channel partners, resulting in a roughly stable total cost stack.

We expect NPV to be consistent with 2019 or slightly better.

To illustrate the effect batteries are having on our cost stack, in 2019 Brightbox hardware costs were about $0.09 per watt of total Sunrun-built installation costs.

In 2020, with new battery capacity more than doubling, we estimate battery costs will become $0.20 to $0.25 per watt of total Sunrun-built installation costs.

If you exclude these additional battery costs, we would expect to see an 8% unit cost reduction.

In the fourth quarter, we deployed 117 megawatts.

Turning now to our balance sheet.

We ended the fourth quarter with $363 million in total cash.

Quarterly cash generation was $22 million after adjusting for safe harboring activity of $27.5 million and the repurchase of common stock of $5 million.

We define cash generation as the change in our total cash less the change in recourse debt and other adjustments, including our safe harboring program, business acquisitions and common stock repurchases.

For the full year 2019, we generated $102 million in cash.

Cash generation can fluctuate significantly due to the timing of project finance activities.

Moving on to guidance on Slide 9. We expect full year deployments to grow approximately 15%.

We expect unit economics or NPV per watt to be at or above last year's level.

In the first quarter, we expect deployments to be approximately 102 megawatts.

Now let me turn it over to Ed.

Thanks, Bob.

Today, I'm going to discuss capital costs, asset performance and recap our investment tax credit safe harbor program.

First, I'm pleased to share that capital costs for residential solar assets continue to decline.

Market data points now clearly support a weighted average cost of capital of less than 5%.

Measured at a 5% capital cost, total net earning assets would be approximately $2 billion or about 30% more than when measured at 6%.

At 5%, contracted net earning assets would be approximately $580 million or about 55% greater than when measured at 6%.

At 5%, our 2019 additions to net earning assets would have been $191 million, 62% greater than when measured at 6%.

And finally, at 5%, our full year NPV per watt would have been about $1.35 or $10,700 per customer, both about 30% greater than when measured at 6%.

In addition, I'm pleased to share that ABS lenders and the key rating agency are beginning to note that our pools are performing better than peers, both in the loan and lease arenas.

This month, we're entering the company's 13th year and across all those years, we have collected $0.99 on the $1 billed.

Although it takes many years for our superior customer value proposition, consultative sales practices, and high construction quality to manifest themselves in long-term observable data versus peers, we believe that this time is soon upon us.

Since December 2018, we've raised $834 million in 3 public ABS transactions.

We expect our next transaction will be ready for market in the second half.

However, given the election, other global events and our strong capitalization, we may delay it or other transactions into 2021 if we believe doing so will result in better execution.

We may also execute a smaller than usual transaction earlier than usual.

The company continues to generate healthy margins which, when paired with our project finance execution, should lead to significant growth both in cash and book value.

Finally, I'm pleased to confirm we are successfully executing our safe harbor acquisition and financing efforts.

Materially all safe harbor products have been received, and we continue to expect we will be able to qualify approximately 500 megawatts of the projects at the 30% ITC level.

We are currently deploying fully safe harbor projects in our direct business, and we are implementing their use in our channel business.

At December 31, our equity investment against this strategy peaked at $27.5 million or less than $0.06 per watt.

This investment represents about 1/3 of the incremental tax credit we expect to receive when deploying this equipment.

This inventory also helps insulate us should any coronavirus-related supply disruption develop.

Finally, year-end cash was $363 million.

Total cash, less recourse debt, adjustments for business acquisition, safe harbor activity and the repurchase of common stock, increased $102 million in 2019.

Turning finally to our pipeline.

Our project debt and tax equity runway each extend into the fourth quarter.

And with that, I'll turn the call back over to Lynn.

Thanks, Ed.

Let's open the line for questions, please.

(Operator Instructions) Your first question comes from the line of Michael Weinstein from Crédit Suisse.

So it sounds like the labor shortages are starting to abate.

And I'm wondering if you could talk about the 2020 growth rate that you're projecting?

And how much of that is impacted due to the ongoing problem as it starts to get solved this year?

And is there any visibility at the end of this market?

Or any visibility towards when it will finish?

So I think we're -- we believe the labor situation has stabilized.

So the -- as we just described in the comments, we went from 300 open positions down to 100, which is pretty normal.

And the -- even 300 opening on the sales side is also pretty normal in terms of ramp rates we've seen in the past.

So we're vigilantly watching because of the tight labor market, but it's nothing that we don't have the recruiters or pipeline to staff.

So it's really not a feature in terms of restraining anything or worry for us for hitting the growth targets for the year.

And you talked a little bit about the cost of installing batteries and the cost of battery acquisition.

Is that -- are those costs expected to go down over time, so that they will start contributing to NPV per watt?

Absolutely.

The -- we are thrilled with the acceleration in battery adoption despite the fact that there's still fairly high cost.

The value proposition is strong, but they shouldn't cost what they do.

So if you just look at the cost curve of the lithium-ion cells and what's happened historically, that's been pretty significant declines.

But the retail price of the actual boxes that we're buying has held.

So there's really still a lot of margin in there in the packaging and the installation.

That is just learning curve, scale stuff.

And with more competition coming in, that obviously will help as well.

So if we look out through the year, we expect to double battery deployments despite the fact we're not projecting any in-year pricing improvements.

So I think that just underscores how prime that this will be when we get to a place where the battery prices start to be realized.

One last question.

Considering everybody is worried about a recession right now.

Maybe you could just talk again -- once again, about where you think the industry is going and where Sunrun -- what will happen to Sunrun in the case of a major recession that might happen?

Yes.

I'll start and, Ed, you will -- I'm sure, have some opinions on this as well.

What's interesting is that, in many cases, we could be countercyclical products.

So we are a savings product for the household.

So sometimes one of the reasons why customers don't choose to adopt solar as they're just waiting or they're not compelled to do it now.

And so in moments where people are looking to savings matter more.

It actually can be a reason to create the urgency and the sales process that we don't always see.

So that's certainly important.

Did you want to add anything?

Sure.

I would say, on the financial market side, I would highlight 2 components.

First, generally, you see significantly lower interest rates manifest themselves in times of recession, which obviously is helpful to the company.

And we are most able to differentiate from a capital -- from a project finance perspective actually in adverse markets.

The most recent example of that was December of 2018 when we priced our asset-backed security transaction inside of the price that peer had achieved at the same time in the market and with about 10 points of additional leverage, I believe.

We are a bellwether and a trusted issuer in that market and tend to see our spread benefits versus peers expand in slightly more adverse conditions.

So I don't have concerns as to capital availability or cost broadly.

But obviously, as you can hear in the script, we want to leave ourselves a little more flexibility than usual as to when we would perform these transactions given our strong balance sheet gives us that flexibility.

I think one other point to appreciate why, again, the residential business we think is so compelling is that we also raise all the capital ahead of signing up any customers.

So we have -- we know the cost of capital before we price and then sign up a customer, which gives you -- you don't get stuck with the backlog of promises.

And finally, I would note, we -- obviously, we've been in business since 2007.

We have performance data from the Great Recession.

We collected $0.99 on the $1 through the company's history, including at that time.

So we feel good about that as well.

Your next question comes from the line of Brian Lee from Goldman Sachs.

Maybe the first one on guidance.

Not to get too bogged down on the shorter-term trends, but the guidance of 102 megawatts for Q1 does put you at 18% growth year-on-year.

So I guess that implies just basic maths and moderation in year-on-year growth through the rest of the year given the 15% full year growth outlook.

Can you maybe talk a bit about the cadence this year?

What's impacting that?

And especially, since it would seem like comps are pretty easy for you in the second half given last year's labor issues.

Thanks, Brian.

Yes.

So the 18%, I think that 18% benefit for Q1 benefited a bit from some of the projects that were in the backlog in Q4.

Now you can only install so much in the winter time, and we do feel like we're caught up on the labor issue there.

So we do believe that we have good visibility for the rest of the year to hit the 15%.

I think we're just trying to just stay -- give ourselves some flexibility and stay disciplined in terms of continuing to try to maintain our unit level economic targets, continuing to really invest in the right channels that we think are really durable long time -- long term.

So that's where the growth rate comes out for us.

Okay.

Fair enough.

That makes sense.

And then maybe just a second question for Ed on the strategy around raising capital, smaller deal earlier in the year, or it sounds like a base case ABS somewhere in the back half of the year.

A lot of volatility in the markets right now, you got the election later in the year.

I know there's a lot of moving pieces you're analyzing.

But generally speaking, cost of capital is pretty darn low these days.

So how much of this is strategizing around that potentially going lower.

I think the markets are pricing in more cuts through the year.

And how much of this is strategizing around just whether or not you need the capital earlier in the year, later in the year?

It sounds like you have visibility out through Q4 right now with the finance that you -- financing you've already secured and have term sheets on.

So trying to understand a little bit of the puts and takes as to what pulls you forward in the year or what keeps you sort of waiting until later in the year in terms of the next ABS here.

Sure.

Great question, Brian.

So we had a robust 2019 in the project finance area, having raised approximately $2.5 billion of capital across all areas of the nonrecourse project finance stack.

And so the -- we start 2020 in a position where we're mostly filling our aggregation facilities, have a lot of undrawn capital.

And so in ordinary course, it probably wouldn't make sense for us to do a major transaction until the third quarter.

If rates continue to be extremely attractive, we might do something a little bit earlier.

It might end up, therefore, a little smaller.

And also, we don't want to commit to entering into a transaction 2 weeks after the election.

So I'm very confident that there are many paths to a great outcome this year, both because, to your point, we have a war chest of undrawn committed capital and multiple opportunities to term things out into the financial markets.

It's just, obviously, now is a unique time to be making full year projections.

And so we're just going to be watching things carefully and doing our best to be executing in as advantageous a fashion as we can.

And Brian, one thing that I would just add to on the growth rate.

The way we think about growth rate is really more around aggregate value that you're going to grow.

And so certainly, the rate matters.

I think that the aggregate amount you grow matters even more.

And if you just look at our performance historically, we installed 2 -- more than the next 2 competitors combined in the year.

So the scale, our scale is quite substantial.

And then if you just look at the per unit value, we also have superior customer unit level economics.

So growth rate matters, but quantity times value matters as well.

So we're really looking at all 3 of those.

Okay.

Great.

Yes, understood.

Maybe last housekeeping one, and I'll pass it on.

The creation costs have kind of jumped around here throughout the year.

Good to see the reduction that you saw in Q4, that was encouraging.

But it did sound like you mentioned there might have been some unique circumstances in the quarter that drove it.

I don't know if I missed them.

But can you kind of elaborate on what those were?

And if any of those do persist into 2020?

Yes, Brian.

Good question.

We did say that there were -- that it was somewhat unusual in Q4 that we did have some mix issues with lower-value projects that had also some lower costs.

We also had a higher-than-trend percentage of cash systems, which also helped the platform services margin in Q4, which was up about $0.05 over Q3.

That, we think, is going to revert back to more where it's been the last -- the previous couple of quarters.

So yes, we don't believe that, that lower rate will continue going forward.

And there were some benefits.

One thing I might add is, it is typical in Q4 that you see lower selling expenses because you have relatively higher installation volume than new customer origination.

You can see that in historical Q4 as well.

Obviously, the installation backlog played a factor for us there as well.

And so that cost -- the sales and marketing cost per unit was especially low as a result of that.

Your next question comes from the line of Mark Strouse from JPMorgan.

I was hoping you could just touch on the California new home mandate, just the changes that SMUD is implementing.

What are your expectations as far as other utilities following suit?

Can you talk about your recent discussions with homebuilders since that announcement?

And then lastly, kind of what are you baking in as far as potential impact to 2020 guidance?

Sure.

First, I would say the biggest impact of the California new mandate is really not going to be the specific megawatts we're going to develop on new homes in the next couple of years, but it's just the normalization of it.

It's the fact that it's going to be more commonplace that people aren't afraid of higher -- harder to sell their home, that people are more comfortable with it.

That is, I think, that -- and the fact that California is committing to a decentralized grid.

Essentially with this policy, that's what they have done.

And when you talk to the people who have made that decision was quite deliberate.

The -- in terms of our discussion with homebuilders, they're progressing very nicely.

We're in discussions or engage with 20 out of the top 30 builders at this point.

It's not going to be a material amount of megawatts this year, as I've continued to say, because builders just have to pull permit this year.

The way you win a builder over is you get 1 community and then you prove yourself, and then you expand.

So it's a little bit of a slower build there.

It's something we're investing in and we want to take more share in, but it's not a huge piece of the current plan.

In terms of this, SMUD decision, we're -- I don't think this is meaningful at all.

First of all, builders like solar, they want to put solar on the roof.

They make more money when they put solar on the roof.

There's this huge misconception that it costs more money.

It doesn't.

With our solar-as-a-service business model, it's no upfront cost to builder.

It's no upfront cost to the buyer.

It's cheaper electricity.

So it is a really attractive product.

And you see leaders like Lennar moved to 100% solar before the mandate was even there because it's a more attractive product.

And the SMUD decision doesn't prevent builders from putting it on the roof.

It just allows for community solar as well, which we applaud.

People who can't get solar on the roof, that's great.

They should -- they can buy it from community solar.

But SMUD is very specific and that they own their own generation as well.

So we don't think of it as a big issue.

I think the facts support, the fact that builders want this and consumers want it, and that will carry the day.

I would -- the one thing I would add, just as a personal prediction is that, as you know, reliability is just a key issue in California, and solar -- and when homes are paired with solar and storage, you get a real key differentiated sale feature that is -- that the homebuilder can use to market the home.

And so I suspect that, particularly in the investor-owned utility service territories, you will see builders start to move more towards storage and solar combined because that is a true differentiated and important factor that is on the minds of California homebuyers today.

Okay.

That's very helpful.

Lastly, you touched on the 5 grid services projects and mentioned you have a strong pipeline.

Just how should we think about new agreements being signed this year?

Are those customers kind of waiting for -- are those partners kind of waiting for more evidence coming out of the existing projects you have?

Or do you think some of those will be signed over the near term?

Some in the pipeline.

Well, the -- I don't think they're waiting for success.

We already have good proof points that we can come in and share.

We're already participating in programs in California and in Massachusetts, where we can show data that it's working, and we're delivering the capacity and aggregating the assets appropriately.

So we have good proof points already.

It's just a slow sales cycle product when you're working with utilities and/or municipalities.

So our strategy is really to -- in parallel, we just want to go direct to consumers and put solar and batteries and get the cost as low as possible to put it on as many households as possible.

That's strategy number one.

In parallel, we're working all these business development relationships with the utilities and grid operators and other operators with the electricity system to open those markets up for us.

So we sort of have those 2 strategies in parallel.

We're not relying on these grid services programs, but they should be a good competitive advantage to us, and they should also help on the customer acquisition cost because part -- a feature of some of these programs is that the utility will, in fact, market to their consumers, which should help us acquire customers in a more affordable way.

Your next question comes from the line of Eric Lee from Bank of America.

Can you hear me?

Yes.

Perfect.

So just to discuss drivers for NPV for '20 relative to '19 with your guidance for at or above.

Can you discuss what's supporting that just in terms of drivers you see?

Sure.

So I think that there are a couple of puts and takes.

So one, the increasing storage percentage is going to bring up that -- the project value as well as a little bit of the cost, but we're going to be able to moderate that cost by installation improvements and other efficiencies.

So whereas the cost stack is going to look somewhat flat overall.

It's actually -- the battery is becoming -- if you look at the Sunrun build cost, for example, the battery is becoming about $0.20 to $0.25 of that cost stack, whereas last year, it was $0.09.

So you have to over -- on the cost side, you would need to overcome that with efficiencies, which is what we plan to do.

So that's where you're going to see the NPV be stable, if not improve for the year.

Got it.

And with the cost reduction efforts you highlighted as well as the ITC safe harbor strategy implemented, how do you think about a long-term sustained trajectory for NPV into '21 and beyond?

From the -- I didn't get the question.

Yes.

Let me just rephrase, I think the question was, keeping in mind changes in the tax credit and also the safe harbor program, what do you see as the long-term trend in NPV.

Was that the question?

Yes, that's correct as well as given the cost reduction efforts you highlighted.

Yes.

So I would say, we don't -- we believe we can sustain the dollar flat through the entire step-down.

So in our investor deck, there's a slide in there that also supports us.

It shows that what we would need to do to maintain neutrality on that is to improve costs by about 4% every year and raise pricing by about 2% each year.

And so we think that's incredibly doable.

So that -- and not only envisions solar.

So it doesn't necessarily envision the fact that consumers are willing to pay even more money to add the battery, and that there's also additional sources of value that can come from the grid services type programs for the battery.

So putting that aside, we think we're well on the path to be able to absorb those declines and have the competitive advantage.

At our scale, right now, we have safe harbor, I believe, more megawatts than any other company.

There's no reason why we won't be able to continue to do that.

So that should be a margin that we're able to capture uniquely.

Got it.

Appreciate it.

And lastly, how do you think about cash on levels for '20 and beyond and maybe use of cash expectations around that?

Well, we're generating cash.

So do you mean use of cash such as buying back the stock or dividend?

Is that your question?

Yes.

Yes, like capital allocation from that perspective, just given the $100 million plus cash in for this year and just expectations for those levels for '20 and beyond.

I know you talked about repeatable levels for about $100 million or so.

Great.

So as I mentioned, we announced the stock buyback program on the Q3 call, which was midway through Q4.

During Q4, we did repurchase $5 million of stock, I believe, at an average price around $13.55 a share.

Obviously, capital allocation over time is dynamic as we consider our own stock price opportunities in the market and our strategy around term outs in the capital markets.

We're continuing to grow into a company even more comfortable balance sheet, and we'll probably always allocate a little cash to that effort as well.

So it's a little bit difficult to give precise guidance.

Other than that, obviously, we do consider it constantly as it's evidenced by the share buyback we did actually perform in Q4.

Your next question comes from the line of Moses Sutton from Barclays.

The decline in creation costs and the associated contracted project value decline, you noted it relates to mix.

Is this geographic mix, perhaps increasing deployments in Texas?

And if so, why would that not continue?

It's both product mix and geographic mix.

And it won't continue because you can just have anomalies based on what went into service that quarter.

And we have pretty good visibility and know where the percentages are coming, where the product mix is coming.

So we have good confidence that, that will not persist.

Got it.

And not sure, I may have missed it, but the $27.5 million safe harbor equity investment, how many megawatts are associated specifically with that exact spend?

500 megawatts.

Your next question comes from the line of Joseph Osha from JMP Securities.

This is actually Hilary on for Joe.

I just wanted to first kind of touch on this doubling of the attach rate for Brightbox.

And if you could provide a little more context for where that demand is going to be coming from and if it's in more of these fire prone areas or if we're going to see those in some of those other markets.

Thank you.

We're very excited about it.

So happy to talk about it.

The -- I would just want to clarify the statement, we said that we were going to double the aggregate number of installations, not the attachment rates.

So I just want to make sure we're clear on that.

Yes, we do believe that California will be a big driver, particularly the areas that have people who have lived through the power being shut off, which is a considerable number of people.

Just to give you some support behind that, if you look at Q2 in the Bay Area, our attachment rate was about 30% and jumped up to 60% and it's stable.

This quarter, it was about in the mid-50s.

So you're seeing just really fast interest in moving towards the battery.

And we have places in Southern California and others, where we're even above 60%.

So it's really -- once our sales force can educate people on this as the product is more appropriate for more home types, we see no reason why the majority of Californians should not switch over to solar plus a battery.

And so that's really what's supporting it.

It's just seeing the momentum market by market.

Okay.

Great.

And then on the retail partner that you already kind of be expanding out in those additional 200 stores.

Just kind of what are the -- what's the time line for that and if we could see a further expansion with that partnership?

Thank you.

Yes, we're excited about big box retail.

We have partnerships with both Costco and Home Depot, as we've talked about in the past.

And the reason we bring it up on the call is we believe that the opportunity to expand into another 20% of stores with this retailer really underscores our differentiation in terms of being able to execute that channel.

So we didn't give any commentary and are still working through staffing plans for that.

But it's a real testament to the fact that this is a differentiated channel for us.

Your next question comes from the line of Philip Shen from Roth Capital Partner.

In terms of -- just as a follow-up to that last question.

The sales and marketing costs have remained stubbornly high.

When we talk to the channel, we hear things like sales guys are printing money.

So can you update us on how you expect to get sales and marketing costs lower?

I know that expansion with big box is an opportunity.

But given you're learning with Comcast and your other channel partners, can you quantify how much you think you might be lower -- be able to lower the sales and marketing costs in 2020?

Or should we expect that line item to remain flat?

Well, I'll -- I've always said, and I will say again that customer value -- the lifetime customer value support our customer acquisition cost.

And we're delivering about over $1 a watt of NPV.

So we're delivering almost $9,000 of value per customer, and that's at a 6% discount rate.

So if we were to market today at 5%, that would be over $10,000.

So we have $10,000 of value.

I would argue that a rational business would go spend another dollar to acquire more customers in order to deliver those type of long-term value.

So that's a -- just an important dimension to remember is that it's not -- the acquisition goes not out of whack with the value.

And in fact, you could maybe argue, you would want to even spend more.

In terms of our -- we believe that we're developing the channels where we're going to be able to have more customers -- I'm going to get it wrong, whether it's pull versus push, but more customers coming to us versus a lot of companies that are really relying on sales people that have more of a hunter type of mentality, which is why they don't really have the marketing spend to actually pull the customers and say you're more reliant on the salespeople.

So that's an important dimension to be aware of.

So when we look at some of our efforts there, we have these relationships with the big box retailers.

We have a pretty sophisticated digital marketing program.

We own lots of web properties that acquire leads from customers.

We are -- have -- are signing up grid services programs where we're going to have utilities market on our behalf.

We have the largest installed customer base to build referrals off of.

All of these things, we believe, are competitive advantages for us that give us the positioning to be able to continue to drive down acquisition costs and not be subject to reliant only on salespeople.

Okay.

We're also hearing in the channel that you guys are more actively wooing dealers.

I know you don't provide a breakout between your dealer and your direct business.

But perhaps, could you speak to the direction of that dealer business, specifically, how does your dealer volume in megawatts trend in '19 year-over-year?

Was it up, flat, down?

And then how do you expect that dealer volume in 2020 to trend?

Again, do you expect it to be up, meaningfully, flat or down?

Sure.

So I said in the script, we expect that our direct business, where it's our salespeople, to grow above 20%.

So that will be growing faster than the channel partner business.

But the partner business is still very important to us.

We think that we can serve that.

We're unique in how well we can serve that population.

We have a nice group of channel partners, none of which is more than 10% of our orders.

So good diversification in there.

And so that will be -- that will deliver nice growth for us, but the direct will grow faster, we expect.

Your next question comes from the line of Jeffrey Campbell from Tuohy Brothers.

First, I just want to clarify the discussion around the permit streamlining.

Does this primarily surround items preparatory to installing the solar system or speeding up attachment of the finished system to the grid or both?

It's typically the -- okay, so the example I gave in Las Vegas has both.

It has an instant building permit and it also has instant utility interconnection.

That's the endgame that we want.

And that's why we've been able to cut out 30 days out of the time to install the customer.

The first step for our SolarAPP program is really to work on the building permit side.

And so what it will do is it will come up with the best practices in an online instant building portal that we can then go to the Asia to say, hey, this was developed with best practices.

Why don't you sign on to this?

It will be more efficient for you and more efficient for your constituents.

And that is -- that's the plan.

So that we're in development of that online portal right now, and we're going to be piloting it with 10 locations, mostly likely in California this summer.

So we're pretty excited about it.

It will be a multiyear effort, but this promises -- this is one of the things I'm most excited about for the prospects of the company because it's undeniable that this can cut out huge amounts of soft costs.

And I believe that with the focus on climate change and the political climate right now, that it will not be a huge lift to get cities and Asia to sign up to a program and a process that serves their people well.

And if it's maybe a little too early to try to quantify what kind of savings we're talking about, is there something that you can...

Yes, absolutely.

So the -- if you look at -- the $7,000 is the bogey we have out there, $7,000 per customer.

That's really the difference between U.S. today and Europe.

So if you look at same cost of hardware, same cost of labor, that's the difference, just given their easier process.

So that would be the bogey.

Probably, we may not get all the way there, but that's what's possible.

Okay.

Yes, so that's significant.

My second question...

And again, more than the ITC step-down.

So you can put that into context.

So while we're talking about the -- what do we do if the ITC falls off?

This could be a huge part of not only compensating, but maybe even improving.

You got it.

I want to ask a question about batteries.

Is it fair to assume that California is a competitive battery market, since it seems like all the suppliers are trying to sell there?

And I was wondering if there are any geographies that are not so saturated that might actually be generating better margins currently.

Or is that not the right way to think of it?

California is still not saturated in any way.

If we think about just some market research we've done in the Bay Area, even a place in the center of solar, many people still don't know that solar and a battery is a superior product to a generator.

I mean just look at -- so if you just look at the increase in generator sales and they're not even available for 9 months, people don't realize yet that solar and batteries are better, more affordable and more durable.

So I think we're still really, really early.

And we are in other markets certainly as well, but we don't believe that we're in the cycle of the industry, where we're really trying to price for like a really big margin.

I mean we're really trying to grow these deployments.

We believe in that.

We believe in the density of creating a lot of density of batteries in the community, so that you can aggregate them and add real value to the grid.

So our vision is we're both capitalist, but we're also really compelled by the need to decarbonize energy.

And the only way we're going to decarbonize energy is to have a huge decentralized piece, and it has to be solar and batteries.

So we're believers in that.

And just one -- couple of numbers to share to underscore that.

I believe, last year alone, something like 6 million customers were involved in a forced blackout because of wildfire risk.

In the state of California, they are probably somewhere between 10,000 and 20,000 installed battery systems.

Right.

So we're -- we've got a lot of ground to cover.

Right.

Well, and as I've tried to make the point, one thing that's great about your installations is that they're all close to load.

And so even as utilities may try to move more and more capital into their own battery farms, I don't think they're going to be able to build them in neighborhoods, and that's going to protect your VPPs, even in that instance.

So, yes, I'm a fan of the model.

Thank you.

Thank you.

We need all of it, certainly.

And if the plan is to electrify, there's plenty of room for us to work together and not to be at odds with one another.

Right.

If I could ask just one last quick one because we've been talking about safe harbor here.

So it sounds like the solar energy systems revs jumped fourth quarter '19 because of the safe harbor capture, and then you said you don't expect that to continue through the bulk of the year.

But I was wondering, is this something that could potentially repeat itself fourth quarter '20?

Maybe fourth quarter '21 is kind of a seasonal grab of the remaining ITC value.

Or do you kind of see this as more as one and done in fourth quarter '19?

This is -- you're talking about just the cash system sales jumping up?

Yes, just the jump up in cash systems, trying to grab the ITC.

I think that -- possibly, I think you probably still will see some of that.

But I think we'll also get more sophisticated at really educating people that they don't have to because we've already saved over the equipment in the past year.

That was new this year.

So I would suspect that you may not see it quite as significant, but there will be probably a little uptick at the end of the year.

Your next question comes from the line of Colin Rusch from Oppenheimer.

This is maybe a simple question, but there's plenty of availability of capital.

There's clearly a major opportunity here and compelling economics.

Why aren't you growing faster?

And then as an augment to that, would you consider inorganic growth as an option given what's the lay of the land looks like?

Well, I would again point out, we added as many customers in 2019 as the next 2 competitors combined.

So our scale is significant here, and we're in an operational business, and we're in a business where we care about -- we really care about the quality of the assets.

I think we're really trying to show, and we do believe that time will really prove out that high quality of build matters, consultative expectations matter on the sale.

And so we balance those, and we think for the long run, and we see where the growth rate comes out.

So we would challenge anybody to get close to the amount of aggregate customer value that we're going to add in the year.

I don't -- I wouldn't imagine that, that would happen.

So I think we're adding a lot of value.

In terms of the inorganic question, we certainly always look at strategic opportunities that are available, but we would never comment on any M&A opportunities or anything of that nature.

Okay.

And then as you look at a growing set of battery assets on the -- in your network.

How many geographies you feel like you could bid into the ancillary services market?

In 2021, 2022, it's similar to what you've done in that few sites.

Yes.

The way we think about that as we sort of have a plan that says, okay, if you just look across the country, what percentage of the population can we cover with grid services like programs.

So hopefully, that makes sense.

And so we think that, over the next 3- to 5-year period, we believe that the majority of the states that we operate in, we will be able to participate in some sort of grid services program.

So we're out there setting it all up right now in order to reap the benefit of that by having the largest solar and battery fleet.

And your next question is from Sophie Karp from KeyBanc.

Just wanted to ask you a question on supply chain.

With everything that we are seeing in, I guess, in China and global trade and considering where the panels come from, is there a chance that we will see some significant disruption in the supply chain?

And how much cushion do you have against that?

Sure, Sophie.

Good question.

First off, I'd underscore that due to our safe harbor program, we have deep inventory in both panels and inverters.

So we are just exceptionally well protected there.

We continue to watch other areas of our supply chain carefully.

We currently see solar manufacturing to be operating at relatively high capacity.

Certainly, we haven't seen price increases, which one would expect to precede any potential supply disruption.

Probably the area where we're tracking most closely is batteries to the extent the challenge were to arise.

Our suspicion is it might be there, although we have no current information to suggest battery supply disruption.

And at this point, we haven't detected any impact in consumer demand either.

Internally, we're obviously also taking steps consistent with guidance from the CDC to minimize any potential transmission risk.

Okay.

I think we will sign off here.

Thanks, everybody.

And talk to you again next quarter.

Ladies and gentlemen, this concludes today's conference.

Thank you for your participation, and have a wonderful day.

You may all disconnect.