慧榮科技 (SIMO) 2019 Q1 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to Silicon Motion Technology Corporation First Quarter 2019 Earnings Conference Call.

(Operator Instructions) This conference call contains forward-looking statements within the Section 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934 as amended.

Such forward-looking statements include, without limitations, statements regarding trends in the semiconductor industry and our future results of operations, financial condition and business prospects.

Although such statements are based on our own information and information from other sources we believe to be reliable, you should not place undue reliance on them.

These statements involve risks and uncertainties and actual market trends and our results may differ materially from those expressed or implied in these forward-looking statements for a variety of reasons.

Potential risks and uncertainties include, but are not limited to continued competitive pressure in the semiconductor industry and the effect of such pressure on prices; unpredictable changes in technology and consumer demand for multimedia consumer electronics; the state of and any change in our relationship with our major customers; and changes in political, economic, legal and social conditions in Taiwan.

For additional discussion of these risks and uncertainties and other factors, please see the documents we file from time to time with the Securities and Exchange Commission.

We assume no obligation to update any forward-looking statements, which apply only as of the date of this conference call.

Today's call will include a question-and-answer session.

(Operator Instructions) I must advise you that this conference is being recorded today, Friday, 3rd of May, 2019.

I'd now like to hand the conference to your first speaker today, Mr. Chris Chaney, Director of Investor Relations and Strategy.

Thank you.

Please go ahead, sir.

Thank you, Amber.

Good morning, everyone, and welcome to Silicon Motion's First Quarter 2019 Financial Results Conference Call and Webcast.

My name is Chris Chaney, Director of Investor Relations.

With me here is Wallace Kou, our President and CEO; and Riyadh Lai, our Chief Financial Officer.

Following my comments, Wallace will give a review of our key business developments, and then Riyadh will discuss our first quarter results and our outlook.

We'll then conclude with a question-and-answer period.

Before we get started, I'd like to remind you of our safe harbor policy, which was read at the start of this call.

For a comprehensive overview of the risks involved in investing in our securities, please refer to our filings with the U.S. SEC.

For more details on our financial results, please refer to our press release, which was filed on Form 6-K after the close of the market yesterday.

This webcast will be available for replay on our website at www.siliconmotion.com for a limited time.

To enhance investors' understanding of our ongoing economic performance, we will discuss non-GAAP information during this call.

We use non-GAAP financial measures internally to evaluate and manage our operations.

We have therefore chosen to provide this information to enable you to perform comparisons of our operating results in a manner similar to how we analyze our own operating results.

The reconciliation of the GAAP to non-GAAP financial data can be found in our earnings release issued yesterday.

We ask that you review it in conjunction with this call.

With that, I will now turn the call over to Wallace.

Thank you, Chris.

Hello, everyone, and thank you for joining us today.

I will first update you on our business, and then Riyadh will review our financials and provide our outlook later on this call.

For the first quarter, our sales from continuing operation of $89 million declined 23% sequentially.

Including FCI, our sales were $95 million and had also declined 23% sequentially.

Earnings per ADS for the quarter were $0.42.

Let me start with an update of our SSD controller.

This quarter, our SSD controller sales were flat sequentially, in line with what we have communicated to you 3 months ago.

As previously discussed, while NAND prices have been falling sharply, which triggered an increase in demand for SSD, the reaction is not immediate, especially with OEM customers, where design, qualification and testing processes will be extensive.

Separately, when NAND prices are falling sharply, module makers face increased financial risk and so become more conservative in procuring NAND flash components and related controllers.

The situation is now changing, procurement of controller for our NAND flash customers is beginning to step up [vis-à-vis] for new OEM programs are starting to move into productions.

Also, procurement of controller for our module maker customers is increasing as they are now receiving NAND flash from their supplier at more attractive terms, which is leading to a rebound in their business activity.

As we have been stating, ever since NAND supply and demand dynamic started shifting from shortage to surplus and NAND pricing has reverted to its long-term declining trend, our business now [files] toward growth, with the benefit of price elasticity of demand.

Approximately 250 million client hard disk drives are still shipping annually.

And over the next few years, as NAND prices fall further, most of these client HDD will be replaced by SSD.

Our pipeline SSD controller projects with our NAND partners expanded further this quarter.

More and more importantly, interest in building SSD especially for OEM application and the interest in adopting more of our controllers in this project from customer has increased meaningfully in recent months.

Based on current design activity, we believe our market share in all the leading PC OEMs will expand meaningfully by the end of this year.

We're excited that our SSD controller order momentum has started to pick up with both NAND flash and module maker customers.

In the second quarter, we expect our SSD controller sales to grow approximately 25% quarter-over-quarter.

While visibility for the balance of the year remain limited, we are expanding pipeline of SSD controller design win, especially for SSD design for PC OEM, to strongly position us for further growth.

Let me, however, caution that while our near-term business momentum has improved, our sales visibility into the second half year remains limited.

While many of our business partners are expecting NAND prices to be stabilized far more moderately in the second half year and demand remain like last year.

NAND flash makers continue to hold elevated level of inventory.

NAND prices remain volatile, and our customer could change business direction quickly.

Our order book for the second half remains unclear and the roughly changing market condition could affect our business, both positively and negatively.

Our enterprise-grade SSD controller remain on track to begin commercial production in the middle of the year.

Commercial shipment of our enterprise SSD controller embedding our Shannon open-channel NVMe SSD are expected to begin midyear.

Additionally, in the second half of this year, we are also expecting to begin initial low volume sale of our enterprise SSD controller to a few U.S. customers.

Turning to our eMMC+

UFS controllers, sales this quarter were down about 45% sequentially.

Our eMMC and UFS controllers have high exposure to the China market.

According to the China Academy of Information and Communications Technology data, first quarter mobile phone shipment volumes in China were down 30% quarter-over-quarter.

Our large eMMC NAND flash customer recently talked about their eMMC sales declining 32% in the first quarter, which is in line with the Chinese data.

Additionally, our NAND flash customer may have exited 2018 with slightly higher controlling inventory, which were consumed in the first quarter.

Based on our current order book, our eMMC+

UFS controller sales growth should be roughly flat in the second quarter.

Looking into the second half of this year, we are expecting strong eMMC embedded memory build activity by Chinese module maker and strong UFS embedded memory sales by our NAND flash customer.

We have a very clear market share leadership with module maker in China for eMMC.

Our module maker customer, who we have been actively supporting for quite some time, have improved sourcing of NAND flash component, specifically for building eMMC and have made good progress.

Securing design wins in Chinese smartphone OEM, we believe our UFS controller sales will grow further in the second half of the year once our upgraded higher-performance UFS 2.1 controller enter production.

Now let me discuss our SSD solutions sales.

In the first quarter, sales of our SSD solutions fell about 40% sequentially, with both our Shannon and Ferri sales declining.

Our Ferri product declined as expected.

Our Shannon product declined worse than expected, as China economics continued to soften and internet companies reduced their operation and CapEx plan.

As many of you know, several leading Chinese internet companies, including Tencent, JD.com and DiDi, recently announced large layoffs.

Our Shannon customers have been reducing SSD procurement plans.

Our open-channel SSD for both Alibaba and another BAT customers are expected to complete testing around the middle of this year and enter production in the third quarter.

For full year 2019, we believe our SSD solutions sales could fall over 10% this year due to China's softened economics and corresponding impact to its internet ecosystems.

2019 remains a challenging year for us.

Nevertheless, we feel better about our SSD controller growth outlook with strong near-term momentum and improving feedback from customer for the balance of the year.

We also feel better about our Chinese module maker customers entering to eMMC market for smartphone, further growth of our UFS controller in the second half of the year and our open-channel SSD project for the BAT continuing to track toward midyear commercial production.

On the other hand, we are feeling less confident about SSD solutions' upside opportunities this year, given [micro] issues in China.

Now let me now turn the call over to Riyadh to discuss our financial performance and more about outlook.

Thank you, Wallace, and hello, everyone.

I will summarize our financial results and then provide our outlook.

Before I begin, I would like to reiterate that our comments today will focus primarily on non-GAAP results unless otherwise specifically noted.

Please note that our non-GAAP excludes FCI; excluded to provide transparency to the performance of our continuing operations.

A reconciliation of our GAAP to non-GAAP data is included with the earnings release issued yesterday.

In Q1, revenue declined 23% sequentially.

Results for our 3 key products are as follows: SSD controller sales were flat sequentially, and I would like to highlight that SSD controllers are now 55% of our total net sales; eMMC+

UFS controller sales declined about 45%; and SSD solutions declined about 40%.

Gross margins in Q1 were flat sequentially at 50.2%.

Gross margins were flat because margin upside from our higher sales mix of higher gross margin controllers were offset by meaningful decreases in gross margins for our SSD solutions and non-storage products.

Controller gross margins were stable quarter-over-quarter.

Operating expenses in Q1 rose significantly to $28.6 million from $27.5 million in the previous quarter due primarily to higher R&D spending expenditures.

Operating margin in Q1 declined to 18% from 26.5% in the previous quarter due to declining sales and higher operating expenses.

Our effective tax rate in Q1 was 14% compared to 8% due to one-off tax benefits in the previous quarter.

Earnings per ADS in Q1 declined to $0.42 from $0.82 in the previous quarter.

Stock-based compensation in our operating expenses, which we exclude from our non-GAAP results, were $4.1 million in Q1 compared to $11.9 million in the prior quarter due to seasonal timing of RSU awards.

We had $281 million of cash, cash equivalents and short-term investments at the end of Q4 -- at the end of Q1, down $8 million from the prior quarter.

Cash flow from operations generated $13.3 million in Q1.

This quarter, we had over $1 million of CapEx, primarily for routine purchases of software and design tools.

In late February, we paid $11 million of dividends to shareholders, the second $0.30 per ADS quarterly installment of our annual $1.20 per ADS dividend that was announced in October of last year.

We did not repurchase any shares this quarter due to limited business visibility with full year.

Now let me turn to our Q2 and full year 2019 outlook.

In Q2, we expect revenue to increase 10% to 20% sequentially, driven by our SSD controllers.

As discussed by Wallace, we continue to have limited sales visibility for the second half of this year, and until we see more concrete procurement forecasts from customers, we do not believe it is appropriate to provide false precision implied by quantified guidance.

Until we have better business visibility, we believe that it is likely that full year revenue could be approximately similar to last year.

Let me provide color on our key products.

In Q2, we expect our SSD controller sales to grow roughly 25% sequentially.

While we do not have clear order visibility for the second half of this year, we nevertheless believe it is more than likely that our SSD controller sales will continue growing strongly through the balance of the year.

We expect eMMC+

UFS controller sales growth to be roughly flat in Q2.

In the second half of this year, we are expecting growth from both our China module maker eMMC sales and our flash customers' UFS sales.

Our legacy Hynix eMMC programs could continue at the current level for the balance of the year.

We expect our SSD solutions sales growth to be flat in Q2 and grow in the second half as our Shannon open-channel SSD solutions enter commercial production.

For full year 2019, we believe that our SSD solutions sales could fall over 10%.

Q2 gross margin is expected to be between 48.5% and 50.5%.

While our mix of higher gross margin controller sales will increase in Q2, we are expecting our controller gross margins to temporarily decline as we share some of our long-term strategic partners' current financial pain.

For the full year, if our product mix remains the same as 2018, our gross margins should remain unchanged year-over-year.

Since our business is dynamic, it is likely that our product mix will change each quarter.

Q2 operating margin is expected to be between 18.6% to 21.6%.

We expect Q2 operating expense to be roughly similar to the previous quarter.

For the full year, we expect operating margin to remain the same as 2018 if gross margin were to remain unchanged year-over-year.

We expect to continue tightly managing operating expenses this year.

Stock-based compensation in Q2 is expected to be in the range of $0.4 million to $0.5 million, and for the full year to be slightly less than 2018.

Tax rate in Q2 is expected to be approximately 15% and for the full year to be approximately 15% our normal tax rate.

This concludes our prepared remarks.

We will now open the call to your questions.

(Operator Instructions) Your first question comes from the line of Karl Ackerman from Cowen.

Riyadh or Wallace, if I could just go back to your full year outlook, it doesn't seem to imply a fairly sizeable ramp in the second half of '19.

You indicated how both revenue and margins would be [close or relatively] similar on a year-over-year basis assuming no product mix change, albeit you just mentioned that product mix will change on a quarterly basis from here.

So I guess, I think one of the biggest variables that impacts perhaps your overall outlook is the open-channel and custom SSD business.

So perhaps you could elaborate on the visibility you have with your both custom SSD and open-channel controller opportunities to the BAT providers not just in the June quarter but also the trajectory in the second half given fairly depressed levels today?

And I guess what gives you confidence that, that segment's growth can advance materially in the second half of '19?

How do we think about the progress and qualification of that program?

So we are working closely with 2 of our BAT customers in China open-channel.

We have been working together for almost -- more than one year.

The progress needs a tremendous effort from both parties, and the hyperscaler need to also a tremendous software development in their side.

We have a tremendous software development to fine tune for different business model on the workload, and we are on track.

Currently, it's in the final stage of qualification in the real workload, and we believe that we will be done by the end of June or mid of June, and product will be moving to production by beginning of third quarter.

So that confidence we have, it says all the -- the [building] is in our hands, but we have to qualify the final stage of the real workload test in the real server systems.

That's helpful.

I guess just for my follow-up, if I may.

On your mobile business, I know you're in a period of transition from growing as a U.S.-based customer while also continuing to service the Korean customers' legacy tech nodes.

At the same time, two other issues appear to be plaguing some in the segment, which are higher levels of controller inventory across the channel, handset supply chain, and perhaps the second being that Samsung wants to aggressively attack the mobile market in the second half.

Given many of these moving parts, how do we think about the linearity of the smartphones controller business for the balance of '19?

Let me try to answer this.

Today, for smartphone, the high end to some of the mainstream will transition from eMMC to UFS.

So that portion, UFS, we're going to grow and depend on one of our major partners from NAND makers.

And we believe we are already showing the momentum last year.

Now they are repacked, moving for UFS 2.1.

That portion, we feel very comfortable.

Regarding for eMMC, so eMMC is going to stay in the value line of the smartphone, also portion of the mainstream and could be in China also in [India].

That portion, I think we have our old partner from Korea that -- because they are transition majority portion into UFS.

So that portion, I think, we were down compared to last year.

However, this year, the price is stable.

However, we see great momentum regarding from China module maker, probably around 6 to 7 customers, they are ramping quickly for the low-end smartphone as well as for set-top box.

That portion will create great momentum for us to grow eMMC, especially from second half this year.

That's also in Huawei and Xiaomi smartphone.

So we feel very comfortable to grow with China module maker for eMMC and grow UFS with one of the NAND maker for second half of this year.

Your next question comes from the line of Craig Ellis from B. Riley FBR.

Wallace and Riyadh, good afternoon, and it's very nice to be back in touch again.

My first question was related to the SSD controller business.

You were clear on outlook dynamics for the second quarter and some of the limited visibility that exists for the second half.

But can you help us understand in the second quarter the degree to which you'd expect the OEM business to perform versus the module maker business?

Do you expect them to grow equally or one materially faster than the other?

So regarding the SSD controller rebound in the second quarter, we feel very comfortable with a high confidence we're going to have a strong rebound about 25% sales revenue growing in the same quarter.

The main reason, because our major project in PC OEM start to ramp.

But also, the module maker, while most of it in China and Taiwan, because the NAND price is relatively stable in the same quarter, they start to procure much more NAND from NAND maker.

Since NAND is still staying in the actual inventory, demand is much bigger than -- supply is much bigger than demand.

So we take advantage all the NAND maker who wants to also sell-through the actual inventory, because we are in the leading position, provide all the different solution for customers, that's why the demand for our side become much higher.

Frankly speaking, we don't have enough resource to support all the projects we have, we can get today.

But we believe in second quarter, both module maker and OEM all grow.

I cannot say equally, 50-50 or 65 -- 55-45, but both are grow much stronger than first quarter.

That's helpful.

And then 2 follow ups to that.

First, regarding last statement about the demand that your customers are placing on you for help with work.

Do you expect to improve your position to meet demand in the back half of the year?

And somewhat related to that, with respect to the opportunity to convert $250 million client HDDs to NAND, how much progress do you think industry will make on that opportunity in this year?

And what do you think is possible next year given where NAND flash prices are relative to HDDs now?

In our view, PC OEM really aggressively try to replace HDD with the SSD.

I think on commercial line, almost 80%, 90% already SSD.

The consumer line, because they have a lot of dual drive, so the -- a lot of the consumer line with dual drive, they will keep at least one HDD, one SSD.

But we also see some with desktop will start to replace HDD with SSD.

But with the NAND price is fall deeply even 2020, we see QLC NAND really become more important, let's say from 2020, almost all NAND maker going to offer 96-layer QLC that is going to trigger another the price decline, price drop, for SSD.

We believe that would be a good momentum for -- we see the SSD moving.

For portable SSD, it's very interesting.

We just start to see the momentum and a lot of customer talk about looking for solution to replace external HDD.

That portion is around 150 million units today, and I think it will be taking another 3 to 5 years to replace external HDD.

And when the majority will require QLC and majority need at least 4 terabyte for external solution.

That will also help NAND maker to really allocate their excess NAND inventory for external solution.

I think that will be the momentum for the next few years to focus on, and that could be a great growth potential for us to sell our controller to NAND maker as well as to module makers.

Craig, let me also add.

When you look at client hard disk drives, they are generally in 3 segments: the computing segment, consumer electronics and external storage.

In the PC segment, the conversion to SSDs are already pretty meaningful, But the conversion to SSD in consumer electronics and external storage are still very low.

So we expect a lot of the replacement of hard disk drives to be happening on these 2 other segments, consumer electronics and external storage, over next few years.

So when you look at our conversion by capacity on the hard disk drive side, over the last few years, it's been one capacity category SSD after another that is becoming cheaper than hard disk drives.

First, 128-gigabyte SSDs and the 256-gigabyte SSDs, and today, the OEM price of a 512-gigabyte SSD is already quite comparable to the price of a hard disk drive.

So even very soon, a 1-terabyte SSD could be as cheap as a hard disk drive, and already, for a consumer, you can go to amazon.com and buy a 1-terabyte SSDs for about $100 or a little bit more than that, and we're talking about retail price.

So the market for SSDs will continue to grow rapidly and the pace of its growth will be largely dependent on how fast NAND prices continue to come down and at what rate hard disk drives continue to be replaced.

That's very helpful color.

And if I could just ask one more, I would just inquire about the OpEx as we look through the year.

My sense is that, as you presented the financial results pro forma without FCI, that all of the operating expense relative to the business is out of the forward-looking model.

But is there any efficiency enhancement or optimization that would come either in the second quarter, in back half of the year related to that divestiture?

Yes.

That's a -- that's a good -- for our divestiture of FCI.

The operating expense for FCI runs roughly $4 million to $4.5 million quarterly, per quarter.

In Q1, it's been -- it was a little higher than normal because of project expenses and other R&D expenditures.

But on a normal normalized quarterly basis, our FCI operating expense is about $4 million to $4.5 million.

So once this deal is done and we expect this deal to be closed in the second quarter, we will be running our business without that.

And so the benefit would be coming through from higher operating margin as well as improvements in our EPS.

Your next question comes from the line of Anthony Stoss from Craig-Hallum.

Couple of questions.

Riyadh, mainly your full year guide to be approximately the same as the $530 million you did in 2018.

Are you assuming nothing from FCI or just the 6 months that it's going to be under your wing in terms of that view of $530 million?

And then also, I'd love to hear a little bit more on the SSD and the client side design wins perhaps for the second half of the year, a percentage basis more than you had in the second half '18?

And then my last question would be, you made mention of shipping embedded SSDs to a handful of U.S. customers, low volume later on this year, can you expand about -- upon what the opportunity might be with U.S. customers in 2020?

All right.

Let me start with the first question and turn over to Wallace for the 2 other questions.

In terms of our non-GAAP revenue for this year, we're talking that about despite our limited sales visibility for the second half of the year, until we have better visibility, it is reasonably likely that our full year revenue could be approximately the same as last year.

And when we're looking at -- when we're talking about that, we're talking about an apples-to-apples basis.

So this year was without FCI being our non-GAAP revenue, and for last year, we'll be also excluding our FCI revenue.

Okay.

Regarding SSD business, I think that from client side, we are experience tremendous demand from customers right now from both NAND partner as well as module makers.

We also have some direct requirements on PC OEM.

So we have a pretty confident in overall client SSD development, also market trend in our product position.

Regarding, you mentioned, our enterprise SSD controller, we're going to ship low volume to a few U.S. customer, that is correct.

Well, we -- we can [hardly] the customer because we are very confident they've been working for almost more than one year and we know exactly what they're looking for, but they're going to try different type of storage application and we definitely looking forward to grow the business in 2020 in both U.S. and China market.

Wallace, do you think that the U.S. opportunity is material to Silicon Motion?

I believe so.

We definitely looking forward to long-term our strategy, that's why put a focus to have a full product roadmap to enter enterprise SSD controller business.

Then if I could just throw in one more question, also more general for you, Wallace.

Have you seen any material changes in your ASPs on the SSD side or any material changes in competition?

I know you highlighted you expect to actually grow share on the SSD client side.

That's a very good question.

As you know, when NAND price declined sharply, a lot of our module maker, they cannot make profit.

And that's why they also become very angry if we can not really share the pain.

So from time to time, we need to reduce our price, our ASP.

But from product mix, the -- because new high-end product coming, that's why we have pretty good confidence to maintain our ASP, because more PCIe.

Also, we can offer PCIe Gen 4 by end of this year.

And so more high-end product coming.

And we also reduce some of our ASP for low end SATA product line.

And that will help to share the pain for our customer.

But we will always see that we can maintain the gross margin, maintain the ASP.

Your next question comes from the line of Mehdi Hosseini from SIG.

A couple of follow-ups.

Riyadh, did you say that the year-end guide that you provided excludes the divestitures?

Our full year revenue for our non-GAAP excludes FCI.

And so when we compare that to last year, we're also looking at it on an apples-to-apples basis.

So excluding FCI.

So you can see this in our financials.

As part of our 6-K earnings release, we provide our GAAP numbers as well as our non-GAAP as well -- and a reconciliation for that.

And there you can see how we exclude FCI for both Q1 and also the prior quarters.

Sure.

And then a couple of follow-ups for Wallace.

Given your rather more constructive outlook for the second half, especially for SSD solution and including open channel, when would you start procuring NAND?

Especially if you -- if you're saying NAND prices are going to go sideways in the second half, at what point in Q2 would you take advantage of opportunity and procure for your own product shipments in the second half?

Yes.

For Shannon business, so for BAT project, our procurement goes through one of SSD manufacturer partner.

This is one of the largest manufacturer partner in the world.

So they will procure the NAND and manufacture for us.

For non-BAT customer, we will procure ourselves.

So every quarter, we procure NAND from the selected NAND partner.

So in that context, would you wait till like June to take advantage of prices before they stabilize?

Or have you already started doing...

No.

We give it a rolling forecast -- every month, we have rolling -- 6-month rolling forecasts to our manufacturer partner.

So price negotiation every 3 months.

We have a price negotiation for the NAND price.

So Mehdi, to put another way, we don't speculate our -- the way we manage our business is not to speculate on NAND prices going up or going down.

It's more -- we're buying for the businesses that we have a quarter [event].

Sure.

Sure.

No.

It's just that over the past 1.5 years, or rather 2 years, due to declining NAND prices, we're speculating demand elasticity, and that hasn't happened.

So how is it different now?

So in -- I think in the past, the Shannon business, because more -- it's very dynamic.

Sometime, our Shannon operation procurement make mistake.

So we have some inventory cannot sell-through.

That's why they hurt our overall gross margin.

And now we believe this arrangement with the Tier 1 manufacturer partner globally, they will help us first to procure very attractive price for the NAND.

Second is we don't need to take a risk for inventory, and that will help us to deal with the BAT major customer.

Yes.

Very clear.

And just one quick for Riyadh.

I'm just trying to better understand how you plan.

Let's say in a worst-case scenario, if the open channel or incremental opportunities were not to materialize, how are you contemplating for your cost into 2020?

And I'm asking this so that we could better understand the worst-case scenario given how you have adjusted an opportunity that could present to you.

So what if they don't happen?

What's the worst-case scenario with the OpEx?

Well, we have our OpEx related to our controller programs, and those we have very clear road map in terms of our projects, our technology, the customers and what we need to do to deliver on those objectives from an engineering and headcount perspective.

And so that road map will continue.

Separately, we have our business plans relating to our Shannon SSD and headcount related to that.

And so the headcount related to that will be dependent on how our business continues to progress.

Let me add comments.

So remember, our enterprise controller support dual mode.

One is open channel, the other is standard NVMe.

So for our engagement with BAT, they needed both SSD solution, I mean, open channel as well as NVMe standard.

For other customers who also can sell same solution and load a different firmware, it's a standard NVMe.

So the development really is no risk for our effort regarding headcount or the capital expense.

Your next question comes from the line of Ari Shusterman from Needham & Company.

So I just want to transition over to just megatrends.

So when it comes to 5G, how do you expect it to affect your business?

Just a little bit of color on that would be appreciated.

Yes.

We don't see the 5G going to impact our business, but we do see the 5G the leading chipset maker like Qualcomm, Snapdragon 865, they're going to move to UFS 3.0 that demand a higher performance.

And our [wire] controller is ready, which is a way for the -- any smartphone maker.

They are going to take for the Qualcomm next-generation their modem the [AP].

But for the data center, they're supposed to have a really new demand for data center, change the overall infrastructure to support 5G because with the higher bandwidth, having more people to use the big data.

But due to the global economy slowing down and also the other factor, now we see that 5G development is a little slower than our original plan.

And data center, especially hyperscaler in China, they also have a more conservative plan in 2019.

And they [got -- that pave the way] for the settlement between the U.S. and China tax war.

So this is a -- potentially impact some of the ramping regarding our Shannon business.

But overall, it's really do not impact for our business.

And one more question.

So the company seems to be significantly exposed to 2 companies, SK Hynix for eMMC and Micron for UFS.

Do you guys expect or have plans to have another major customer in either of these 2 spaces in near or long term?

As I said, we do see the China module maker start to grow strongly, especially for eMMC.

But we have as 7 -- 6 or 7 major module maker that are winning for smartphone, also set-top box, and they are growing rapidly.

And we believe the ramping will become more meaningful.

It would be from second half.

And that portion will continue to grow because the eMMC going to stay with the low-end smartphone.

All the mainstream high end will then move into UFS.

But these customers eventually also will adopt UFS.

So we feel more balance for our business for mobile, and we feel much comfortable also, I think, that we have all Micron's UFS socket that we believe they're going to grow, especially in second half this year.

Our next question comes from the line of Donnie Teng from Nomura.

My first question is regarding to the full year outlook.

Although the visibility is quite low, but as you mentioned about it is likely -- there's still a chance to approximately similar to 2018 level.

And you just mentioned about the SSD solutions sales to decline over 10% year-on-year this year.

I'm just wondering if we can really meet a similar sales level as last year.

What kind of growth magnitude on client SSD controller and eMMC controllers will be?

And I will have follow-ups on the different business.

Yes.

I think, as we said, during the China -- hyperscaler have a more conservative plan for SSD solution.

We believe this year we're about 10% decline compared to 2018.

Now for SSD controller, we have a much stronger position, although Q1 is a little disappointed.

But we do see the rebound in Q2.

We also will see probably even stronger rebound in the second half this year because the growing PC OEM, not just the NAND maker.

We also assist probably 3 to 4 module maker and the PC OEM program, and that will be really ramping quickly.

That's why we say we have the confidence this year the PC OEM market share we're going to see the increase compared to 2018.

For mobile, we believe overall, we'll stay flat compared to 2018.

Although second half may be down and -- but we see that largely similar.

That's why overall we see -- that's why we -- and if you remove the FCI portion, we probably fairly is similar to 2018.

Got it.

So for different business segments, so for client SSD controller, I do a simple math.

It's like yours -- your first half client SSD controller sales will grow like close to 20% year-on-year.

And just wondering, what kind of sales growth for the overall industry.

Do you have any idea about this year?

And how should we expect SIMO's client SSD sales growth?

To be higher or lower than the industry average?

Donnie, let me step in and provide the caution that Wallace had provided earlier in his prepared remarks.

We're very confident about the programs that we have.

We are -- we still have very limited concrete visibility into the second half of the year.

We have all the programs.

We know where our products are going to be going into with PC OEMs and to which channels.

But the same time, we are still not able to have concrete order forecast for the second half year.

And so when we talk about our full year outlook, we're not saying that we're going to be growing.

0% flat.

We're talking about that based on the momentum and what we see -- what we're seeing right now, there is a reasonable likelihood that our sales this year could be similar to last year without providing even more precision on a quantified basis.

And for eMMC-related sales, did you just mention about that it is likely to be flat as last year?

So if we make an assumption behind this, like can we assume that the Korea major customer's orders are still declining but can -- 100% offset by the new customers like China module makers and maybe U.S. customers in second half?

Yes.

I think that based on current forecast, it look like the other Chinese module makers, there's eMMC demand can offset the decline from Korean customer.

Got it.

Got it.

And then...

[Also our UFS] business.

Got it.

And my one last question is a housekeeping.

So I'm not sure if I listened correctly.

So you said that you will have low-volume sales on enterprise controllers to U.S. customers, right?

Small volume.

Yes.

Correct.

Second half this year.

Your next question comes from the line of Charlie Chan from Morgan Stanley.

So first of all, I want to ask about the enterprise solution business.

So you mentioned that you outsource those NAND flash procurement to your kind of Tier 1 manufacturing partner.

So my question is whether those NAND flash costs will be counted in your revenue when you sell those solution to your customers.

Charlie, you would -- the NAND flash costs would continue to be part of our cost of sales.

Okay.

Okay.

So I'm wondering because you mentioned that -- you're confident that the open channel solutions can be qualified by customers on time.

My question is that do you expect that the solution business revenue overall can recover to previous peak of revenue?

I remember it could be like USD 35 million to USD 40 million during a peak.

Do you think we can get to that revenue level in third quarter this year?

We're -- Charlie, we're a little hesitant in putting a number and attaching a number to a quarter at this point in time given that we still have a lot of work to do.

But we feel fairly confident that we're tracking towards completing our testing by middle year and immediately afterwards going into commercial production.

But overall, our SSD solution this year of our Shannon plus Ferri products will be down by at least 10% compared to last year with the primary driver not because of our open channel but because of what we're seeing in China with the softening economic conditions and the -- our Shannon customers dialing back and postponing their procurement in order to derisk their business outlook.

And so this is leading to lower levels of procurement for the full year based on what we're seeing today.

So Charlie, let me add one comment.

I think just on the math, you need to understand is the enterprise, the ASP declined almost 50% compared with last year.

So even we ship the same big growth, a 20% big growth, our revenue for SSD client device solution will decline 30% from last year -- you understand that?

Yes.

Yes.

That's why -- it doesn't mean we are losing business, it's just the total dollar declined compared with last year.

Okay.

Okay.

Yes.

So that's fair enough.

So let's -- follow on this part of the question.

So I remember over the past 2 years there were some other dynamics.

For example, does a standard enterprise SSD price become very competitive and they kind of jeopardize your kind of customized solution?

And also, do you discount the kind of a weaker total CapEx in the second half?

Or you're assuming a recovery of those China BAT sort of CapEx in second half as well?

I can -- I think I can only say I cannot speak with the China BAT.

But I think open channel, [we're definitely] corporate strategy for both the 2 -- at least 2 of the 3 BAT customers.

That's why there'll be long-term strategy focused on open channel.

But however, this year, due to various factors, the China hyperscaler will becomes more conservative than their procurement plan compared to original plan around late last year.

Now also, due to the ASP declined sharply, I think one of the leading NAND maker make a very aggressive pricing -- that's why the -- overall, the price declined sharply for enterprise SSD solutions.

So even we sell the same big amount compared with 2018, the dollar -- the sale revenue will decline.

Okay.

Okay.

And my second question is regarding your client SSD market share.

So I think there are some news on some reporting about competition from Asian competitors like Phison and Realtek, right?

So I just want to get kind of an assumption about, first of all, the kinds of the industry growth for this year according on your opinion.

And secondly, your market share assumption.

And lastly, the ASP erosion in SATA and the PCIe, respectively.

So first of all, as we mentioned, we are not losing market share, but we are naturally growing market share this year.

Secondly is from time to time, we do see competitor use a very aggressive pricing to approach our customers.

But from time to time, we have to consider whether we need to share the pain if the price drop is reasonable.

So based on different condition, we will discuss total bundled package with the customers.

But so far we see -- we maintain very strong positioning on all module maker, and we're gaining more OEM program from the top 5 PC OEM customers.

That's where we are.

Regarding the PCIe and SATA, I think the -- we do see PC OEM moving toward more for the PCIe, SATA portion, they're going -- it's not going away, but the total percentage is declining.

And we also see in China the retail also start to adopt PCIe.

And all our new projects in PC OEM, they are almost 95% PCIe.

But for the channel, we still see SATA demand steadily, stably from both China and U.S., also Europe.

And I think SATA got a long tail.

However, PCIe is going to grow strongly for PC OEM, also as well as the retail.

Charlie, let me also add to your question about the potential competitor that you had mentioned.

Late last year, a few of our module maker customers that used some of our competitors' controllers, these are companies with a shorter track record, and these are for projects with a fairly low volume.

So what we're seeing is some of these projects are already ramping down, and so we don't feel that this is a threat at all to our business position.

And as you know, controllers are hard to do well.

And in the past, we've also faced a lot of competitors who wanted to come in this space and most of them have disappeared, if not all of them, because they have not been able to stay relevant year after year.

As you know, our customers have a choice -- can make choices.

For them, it's all about do they want to buy a Yugo or do they want a Mercedes or Toyota.

Yes.

Okay.

So long question short, I mean, for this year, I remember you mentioned that client SSD revenue can grow more than 20% year-on-year for 2019.

And what was the kind of your new assumption now?

We also have a -- I think from the momentum in Q2, we are confident.

But we have a very low visibility for second half this year.

We definitely we can grow client SSD controller business this year because we're the only one major focused company for all NAND maker.

When they want to move their actual inventory, they'll come to us.

Your next question comes from the line of Sujeeva Desilva from Roth Capital.

Wallace, Riyadh, I just want to make a clarification here.

You talked about the China BAT customers, that open channel is strategic for them.

And you also talked about them having headcount reductions.

I want to know if any of the open channel projects that they had at their firms were impacted by those headcount reductions or whether they remain strategic.

Look, we say headcount reduction, it's for Tencent and JD.com and DiDi.

Okay.

Okay.

Got it.

That's good.

That helps.

And then UFS ramp, is that more of a second half of ' 19 ramp or a 2020 ramp?

I know you're going to hit more when UFS becomes the volume part of the market versus the premium.

So when does that cutover start to help you guys in your revenues?

No.

We already start to ramp the UFS from last years.

But just the first half looks like the momentum is -- the demand is slow.

But we see that currently on the forecast, we're going to ramp rapidly from second half this year because there are more new model coming from China as well as Korea.

Suji, let me also add our UFS last year were already 10% of our overall UFS plus eMMC sales revenue.

Got it.

And are these models, guys, premium or mainstream models for UFS at this point?

Some are mainstream, some are premium.

Yes.

Okay.

So you're starting to see mainstream.

Okay.

my last question is about PC's SSD attach rates.

Where do you think that trends in '19 versus '18 given that SSD pricing has come in that maybe the PC OEMs want to be more aggressive?

Where are the attach rates trending year-over-year?

Well, I think the consensus is around 20%, but we really don't know like HDD going to stay with some model because it's a dual drive and PC OEM are not going away to renew the HDD.

But that's a dual drive.

But we do see there's more different type of all-in-one PC coming.

So there's a lot of different variety needing storage, and we do see the momentum from the PC OEM.

They plan also for 2020 and 2021.

So let me also add.

The adoption of SSDs in PCs are reasonably high already, but there are still a lot of opportunities to displace the remaining hard disk drives in the PC market.

But you also have other client device segments that are largely untapped by SSDs, and these include consumer electronics and external storage.

Your next question comes from the line of Gokul Hariharan from JP Morgan.

My first question is on -- could you give a little bit more color on what is your client SSD controller mix right now versus PC OEMs and module makers?

If you could give that mix?

As well as could you give us some idea about what percentage of the market that you're serving in terms of client SSD is still in SATA?

And how much of it is moving to PCIe as we get into second half of the year since you had mentioned that almost 95% of your new projects are for the PC OEMs, so based on PCIe?

That was my first question.

So regarding the ratio between OEM, module maker, we'll say we're about 50-50, PC OEM around 50% and the module maker also around 50%.

Regarding SATA versus the PCIe, we definitely have a much stronger demand for PCIe from PC OEM projects.

On the channel side, I believe 80% is TM and PM SATA.

We'll see the transition -- we believe by end of 2019, PCIe SATA will be 50-50 while PCIe will be a little higher.

Okay.

Understood.

Just one more question on your enterprise controller solutions that you mentioned that you'll start in small volumes for U.S. customer.

Are these kind of customers also hyperscale data center customers?

And are these engagements also related to open channel flash?

Or are they more traditional enterprise or data center SSD controllers?

So I will say we definitely try to approach all the possibility.

But currently, we are engaged with U.S. open channel customer.

But the volume shipment in second half 2019 is not open channel.

It's a standard PCIe NVMe.

Okay.

And is that for data center customers?

Or is it for some of the -- your similar enterprise OEM kind of customers itself?

Or primarily for your customers on the...

It's for enterprise -- it's for leading enterprise customer.

Okay.

One last question.

In the previous years, [Ben], I think, obviously stock price has [connected] quite a bit.

Now In the previous years, you guys have exercised buyback options.

And obviously, you're still fairly cash rich at this point in time.

Any thoughts in terms of potential buyback of the stock given that the stock price is pretty much similar to where you had exercised some of those options in the past?

Well, we have a share repurchase program for 20 months -- 24 months -- authorization for 24 months, and this was given by our board late last year.

So our normal objective is to buy as much as cheaply as possible to maximize earnings accretion.

But our related principle is in order to repurchase, we need to have good visibility into our business.

So with good business visibility as well as our visibility into cash flow generation, and with the objective that if visibility is good and valuation -- our current valuation is low, we should buy good -- buy a lot of shares.

But if visibility is poor, we're going to be stepping away from that because we will then have -- we will not have clarity in terms of our cash flow and, therefore will not be buying our shares.

And so currently, the situation we're facing is business visibility is still quite limited, and so we're going to be temporarily being a lot more cautious about approaching buybacks.

That's all the time we have for questions today.

I'd now like to turn the call back to Mr. Wallace Kou for his closing remarks.

I would like to thank all of you for joining us today and your continuing interest in Silicon Motion.

We will be attending several investor conferences in Asia and U.S. during the second quarter.

Detail of these events will be available on our website.

Thank you for joining us.

Ladies and gentlemen, that does conclude our conference for today.

Thank you for participating.

You may all disconnect.