Power Integrations Inc
NASDAQ:POWI
Earnings Call Analysis
Q4-2023 Analysis
Power Integrations Inc
The company reported a revenue drop to $86.2 million in the fourth quarter, pivoting the narrative on a challenging quarter but with an optimistic outlook for recovery. Management anticipates revenues to rebound to around $90 million in the first quarter of the next fiscal year, with a margin of error of $5 million either way.
Gross margin, a key indicator of profitability, saw a minor contraction to 52.7%, decreasing by 60 basis points from the previous quarter. This slight dip was attributed to lower manufacturing volumes, which are expected to recover, alongside operating efficiencies and a more advantageous mix of industries served. The expected improvement in the exchange rate and higher manufacturing utilization rates signal better margins ahead.
The company displayed effective cost management with non-GAAP operating expenses for the quarter being $40.3 million, down $1.5 million sequentially. Despite this reduction, expenses are projected to climb moderately to approximately $42.5 million in the next quarter due to increased headcount, resumed taxes, and higher employee benefit premiums. A high single-digit growth in non-GAAP operating expenses is foreseen for the full year.
From a capital allocation perspective, the company generated $16.3 million in cash flow from operations during the quarter and saw inventory days increase to 344, mostly due to lower revenues. A strategic move, representing confidence in its stock, was the repurchase of 680,000 shares—an investment of $47.4 million, which is considered opportunistic by management. Additionally, $99 million was returned to shareholders as buybacks and dividends, showcasing a commitment to returning value even as they are managing capital expenditure prudently at $21 million, nearly 5% to 7% of sales.
Looking forward, the management expects to see a decrease in channel inventory which sets a foundation for potential sequential revenue growth in the upcoming June quarter. The forecasted non-GAAP gross margin for the next quarter is around 52.5%, with a favorable outlook for improvement driven by beneficial currency rates and higher output in manufacturing, leading to better gross margins later in the year. The non-GAAP effective tax rate is projected to be about 7% for the first quarter and throughout the year.
You don't have any saved screeners yet
Thank you for standing by, and welcome to the Power Integrations Q4 Earnings Conference Call.
I would now like to welcome Joe Shiffler, Director of Investor Relations to begin the call.
Joe, over to you.
Thank you. Good afternoon, everyone. Thanks for joining us. With me on the call today are Balu Balakrishnan, Chairman and CEO of Power Integrations; and Sandeep Nayyar, our Chief Financial Officer.
During this call, we will refer to financial measures not calculated according to GAAP. Non-GAAP measures for the fourth quarter exclude stock-based compensation expenses, amortization of acquisition-related intangible assets and the tax effects of these items. A reconciliation of non-GAAP measures to our GAAP results is included in today's press release.
Our discussion today, including the Q&A session, will include forward-looking statements denoted by words like will, would, believe, should, expect, outlook, forecast, anticipate, and similar expressions that look toward future events or performance. Such statements are subject to risks and uncertainties that may cause actual results to differ materially from those projected or implied. Such risks are discussed in today's press release and in our most recent Form 10-K filed with the SEC on February 7, 2023. Finally, this call is the property of Power Integrations. Any recording or rebroadcast is expressly prohibited without the written consent of Power Integrations.
Now I'll turn it over to Balu.
Thanks, Joe, and good afternoon. As expected, fourth quarter revenues were lower as a result of soft demand and elevated supply chain inventories, and we expect first quarter revenues to be about flat sequentially, reflecting these continued headwinds. However, while channel inventory is still above normal, it fell by more than a week during the quarter as sell-through exceeded sell-in by a considerable margin. In dollar terms, we are at our lowest level of channel inventory in 2 years, and we expect further decline in Q1. We are especially encouraged by lower inventories related to the appliance market, which accounts for the bulk of the consumer category.
Distribution sell-through for consumer was up sequentially in Q4 and far exceeded sell-in, bringing the channel inventory back to normal in terms of weeks and to the lowest level in 8 quarters based on dollars. End customer inventories have also improved considerably over the past several quarters, and we are seeing an uptick in bookings from customers that are largely dormant throughout the last year.
While appliance demand is clearly being hampered by the downturn in housing, our consumer revenues in 2023 were below even the pre-COVID levels of 2019, suggesting that we are shipping well below end demand and could be poised for a recovery in 2024. In fact, we expect consumer to lead the way as we begin to see overall sequential revenue growth beginning in the June quarter with a more meaningful improvement in the second half of the year. While 2023 was a difficult year with revenue down more than 30%, there were pockets of growth in several areas that are key to our long-term growth strategy. Our high-power driver business had a second consecutive year of growth even as the broader industrial category was down almost 40%. We had an outstanding year in terms of design wins in high power with a projected annual revenue value of the design wins up more than 70% from the prior year.
Renewable energy was the major driver of that growth with significant wins, not only in the utility scale, solar and wind markets, but also in the adjacent high-voltage DC transmission market. On our July call, we announced a major multiyear award for an undersea link connecting North Sea wind farm to the mainland. In Q4, we received an initial multimillion dollar purchase order for the design as project prepares to ramp up later this year. Another bright spark in 2023 was India, where revenues increased year-over-year and are approaching 10% of total sales.
This is not just a result of manufacturing moving out of China, but also a rising level of in-country design and production for the domestic market including a rapidly expanding middle class and modernizing infrastructure. We are participating in a number of ways with significant design wins in 5G fixed wireless, smart utility meters and appliances, including ceiling fans, which are converting to brushless DC motors, utilizing our BridgeSwitch motor drive ICs. We also have a strong pipeline of design opportunities in electric transportation in everything from 2-wheelers to buses and locomotives. Speaking of which, we made tremendous progress in our automotive business in 2023, racking up wins and expanding our design pipeline in high-voltage EV applications such as drivetrain emergency power, 12-volt battery replacement and micro DC-to-DC converters. Our automotive qualified products are extremely well suited for these applications, which not only require high efficiency, but also benefit from the reliability and the space savings of our low component count designs. Eight car brands are now shipping vehicles using InnoSwitch or scale iDriver in traction inverter applications. Meanwhile, our pipeline of EV design opportunities grew by more than 80% in 2023 with sample stage designs at various levels of progress across all regions and most major Tier 1s and OEMs. We expect several such designs to start production later this year.
Another 2023 success story was GaN, not only in terms of revenue growth, but also key technology breakthroughs, including the introduction of 900-volt and 1,250-volt GaN switches. While other suppliers are limited by capabilities of foundry-based GaN technology, we designed our proprietary GaN to support higher voltages, and we expect to announce the next step on the road map in the near future. GaN has significant cost advantages over silicon carbide in the voltage and power ranges that it can address. And we expect the overlap between 2 technologies to increase over time as we further advance our technology and bring out more system-level GaN products.
As indicated by recent M&A activity, market participants are recognizing the potential of GaN to be a transformational technology in power electronics with the huge opportunities in markets such as automotive, data center, appliances and mobile devices. Proprietary technology and know-how in high-voltage GaN or scarce assets and Power Integrations has more than anyone else in the market.
Our latest GaN product introduction came last week with InnoSwitch5-Pro, which will ship with a choice of 750 or 900-volt GaN switch. InnoSwitch5 is a shining example of our system-level approach to power conversion technology, marrying the efficiency of GaN with a novel control scheme that implements high-efficiency zero voltage switching or ZVS with only a single GaN switch versus 2 switches required in typical ZVS designs. The combination of GaN and ZVS delivers efficiency of better than 95% and with very low component count, enabling exceptional power density for high-power chargers up to 220 watts. We demonstrated this capability with the new reference design showing 140-watt USB PD charger with a volume of just 4.2 cubic inches less than half the size of standard 90-watt notebook adapter with over 50% more power.
To conclude, while 2023 was a challenging year, and the current down cycle has been severe. We have weathered it by sticking to our playbook that has served us well in past downturns. That includes building wafer inventory to protect our dedicated foundry capacity and to be ready for a strong upturn. It includes buying back stock at opportune moments as we did in Q4 and it includes true debt expense control, along with continued investment in GaN, EVs, motor drives, renewable energy and the India market. And while the slope of recovery is uncertain, we see good indications that sequential growth will start in the second quarter with a better second half to follow.
With that, I'll turn it over to Sandeep for a review of the financials.
Thanks, Balu, and good afternoon. Fourth quarter revenues were just under $90 million in the middle of our guidance range, while non-GAAP earnings were $0.22 per diluted share, above the level implied in our guidance, thanks to lower operating expenses and a $0.04 tax benefit. For the year, revenues were down 32% to $445 million, while non-GAAP earnings were $1.29 compared to $3.29 a year ago. While this was a challenging year from a financial perspective, we believe we have managed the business prudently while staying focused on the long term as always. We held non-GAAP expenses growth to less than 2% in spite of high inflation without reduction in employee compensation of benefits. In fact, we gave normal raises last year and continued our practice of paying an above average portion of the cost of benefits despite extreme price pressures in the insurance market.
As Balu noted, we maintained this expense discipline while making necessary investments for long-term growth, including our automotive efforts, our expanding presence in India, continued development on our GaN technology and the products that will double our SAM by 2027.
Now I will quickly recap the Q4 results and the outlook, and then we will open it up for questions. Revenues for the quarter were just under $90 million, down 29% from the prior quarter with all 4 end market categories sequentially lower. As expected, communications was down the most with a decline of about 40%. As we noted on last quarter's call, we had significant restocking in the previous quarter by distributors serving the Chinese handset supply chain.
We also saw a steep decline in Q4 related to an inventory correction at a non-Chinese handset customer. The correction also affected tablets, which drove a decrease of about 35% in the computer category. Industrial and Consumer revenues were each down about 20% from the prior quarter, again driven by elevated supply chain inventories and soft demand.
Revenue mix for the quarter was 35% Industrial, 29% Consumer, 27% communication and 9% computer. Distribution inventory ended the quarter at 10.5 weeks, down more than a week from the prior quarter as well as sell-through exceeded sell-in by about $12 million. Non-GAAP gross margin for the fourth quarter was 52.7%, down 60 basis points from the prior quarter, driven mainly by lower manufacturing volumes, partially offset by a more favorable end market mix. Non-GAAP operating expenses for the quarter were $40.3 million, down $1.5 million sequentially and below our forecast as we continue to manage spending carefully while prioritizing investments in our long-term growth.
As noted earlier, we recognized a tax benefit in the fourth quarter from the reversal of FIN 48 reserves, which contributed about $0.04 to the non-GAAP earnings by share of $0.22. The FIN 48 reversal had a larger effect on GAAP results, resulting in a negative GAAP tax rate for the quarter and GAAP earnings of $0.25 per diluted share. Weighted average share count fell by about 0.5 million shares to $57.3 million, driven by repurchases.
Cash flow from operations for the quarter was $16.3 million. Inventory days were at 344 at quarter end, up 114 days from the prior quarter, driven largely by the lower revenue number. As a reminder, the bulk of inventory is held in wafers which combined with the fungibility of our products across customers and applications result in minimal risk of obsolescence. We expect inventory days to trend down as revenue starts to recover in the June quarter.
Our largest use of cash in the fourth quarter was share buybacks. We repurchased 680,000 shares during the quarter, well above 1% of our shares outstanding for $47.4 million. The average price per share was just under $70. Other uses of cash during the quarter included $6 million for CapEx and $11 million for dividends. For the year, cash flow from operations was $66 million, while CapEx was $21 million, just below our model of 5% to 7% of sales. We returned $99 million to stockholders in the form of buybacks and dividends during the year, more than twice our free cash flow.
Turning to the outlook. We expect revenues for the first quarter to be $90 million, plus or minus $5 million. We expect sell-through to again exceed sell-in, resulting in another decrease in channel inventory and clearing the way for sequential growth in the June quarter. Non-GAAP gross margin for Q1 should be approximately 52.5%. I expect a rebound in gross margin in the June quarter driven by the favorable yen exchange rate and higher manufacturing utilization as we begin to convert more wafers to finished goods.
I also expect a better mix as the industrial and consumer markets begin to recover. Non-GAAP operating expenses should be around $42.5 million in the first quarter. The increase from the fourth quarter reflects headcount increases, resumption of FICA taxes and higher premiums for employee benefits. For the full year, I expect non-GAAP OpEx growth to be in the high single digits. Finally, I expect the non-GAAP effective tax rate for the first quarter and the year to be around 7%.
Now operator, let's begin the Q&A.
[Operator Instructions] Our first question comes from the line of Christopher Rolland with Susquehanna.
I guess, maybe looking into next quarter, if you guys could kind of illuminate inventories, how you see the dynamic per end market playing out? And I don't know if you want to force rank those. What's -- versus flat, what's up or down would be great.
So as we had indicated, the overall channel inventory came down about 10.5 weeks in, but the highlight was that the consumer and the computer segment came down to what I would call normal levels. We are still a little elevated in industrial. And considering that the revenue is it's at normal levels based on this current level of revenues, which bodes really well for what it looks like, as I've mentioned about the consumer segment in the second quarter and further going forward. So we -- I expect the channel inventories to further come down and be below 10 weeks in -- at the end of the first quarter.
Yes. No, I was actually talking about the end markets. Where do you have the most inventory, or how do these dynamics play out into March, and then force rank versus flat, which might be up, and which might be down in March, sorry.
Yes, Chris, that's what I was talking about, that the channel inventory, that's where our guidance is what's in the channel and what's as a result, is probably getting impacted with the customers. that consumer and consumer category are down to our normal levels in the channel, which means things in the end state has become normal, while as industrial is still very elevated and above the level and that's as a result. What I also tried to mention was that considering it's at a such low level, and if consumer and computer are already at our level in the channel at that level as the revenue increases, it's actually will become even much below our level.
Yes. You have to remember, by weeks, if you are normal, in dollar terms, we are well below normal because our -- the weeks are based on a $90 million revenue. As revenue grows, it becomes -- it is really problematic for the customer if it is that low. They have to buy products, which is good for us.
Okay. Yes, I hear you. Maybe something on capital allocation. So nice buybacks in the quarter, Sandeep, I think if this chip thing doesn't work for you can be a day trader. But what are you guys doing on the capital allocation front, more buybacks kind of at this level? If you could and -- or would you entertain something like M&A? Have you looked at that? How are you feeling on this, particularly for the cyclical bottom?
Chris, as we have talked about, we have a 4-pronged approach to Capital Cam. The first being the internal investments. And as you can see, we are investing in working capital as well as you will see from the expense investments that we talked about high single digit for next year. We've tightened our belt over the last few years, and we need to make more investments. So you're going to see, even though we held back in our head count in 2023, you will see us see quite a bit of investments in headcount to support the initiatives that Balu, and I have talked about, whether it be India, whether it be automotive, whether it be motor control, and to add in GaN.
The second thing is M&A and I think as we've talked about, we are always looking at that angle, and we always will look for either technologies that will complement us or which can help us expand our SAM. So that is, I think, something that we do all the way time and we are continuing to look at that. The buybacks is really an opportunistic story. And I think as I presume you like the buyback that we did last quarter and the price of it. We always do that, and it's not that we are rigid, it's opportunistic, and we tend to buy as the stock comes back, and I think we are not going to change that.
And the last one was the dividend, and you know a quarter ago or so we upped the dividend again. So I think we've been following the 4-pronged approach now some things happen more frequently than the others, but it is something that is looked at every quarter with the board, and I don't think anything will change in that aspect of it.
Our next question comes from the line of David Williams with The Benchmark Company.
First, it sounds like you're expecting June to be a bit better, and we've heard that from others. I guess maybe if you could talk about what's given you -- maybe on these inventory side, but what's giving you that confidence? And maybe if you could speak to the booking velocity across the end markets and kind of how those are trending and maybe what you've seen thus far into the year?
Well, there are several reasons that we feel good about Q2 being a growth quarter. First one is the inventory will be less of a headwind. Now it's getting closer to normal. And the -- as a result, we expect that this year would be -- we'll have a normal seasonality up from Q1, especially in consumer because of the air conditioning consumer could even be stronger than normal because of low levels of channel inventory that we talked about.
We are at normal levels of weeks based on -- but based on a very low revenue number. And as you know, air conditioning peaks in Q2, and we have a really good exposure to the air conditioning market. Also, we see some recovery in the cell phone customer where who -- if you remember, they canceled some orders in -- starting in Q3. And I think that inventory correction will be completed in Q1. We did have a slight uptick in Q1 but we expect to come back to a higher level in Q2. And that also helps us in terms of growth in Q2.
The industrial market, which has been very soft for a long time, we think we'll begin to recover sometime in Q2. So that will also contribute, we believe. Probably the last one, which is probably the most important is that our order trends are improving. January was our best month of booking since last spring, and February is also off to a very good start. Of course, the next 2 weeks will be soft because of the Lunar New Year, so we'll see how things will come back after the holidays. But all of the trends support Q2 being up from Q1.
Okay. Great. Very great color there. Certainly appreciate that. And then secondly, I wanted to see if there's maybe a way to think about your design wins. You've had a very healthy pipeline through the downturn. And is there a way to kind of think about that contribution maybe in the next 12 to 18 months from the design wins that you've won in this more recent period.
You can't directly correlate because the design wins come in many different shapes and sizes. And also, you have to remember, some of them will be replacing designs that are in production, but going end of life. So you can't directly correlate it. But directionally, it's a very good trend, and we are really excited about that. If you look at literally almost all markets, except maybe cell phones, which is, as you know, is not a growth market. We are seeing significant design win increases that will bode well for the future.
So the design wins will convert into revenue in the next 1 to 3 years, depending on the market. So that bodes well for the market. Of course, all of this is tempered by the macro situation. So we are waiting for macro to turn around. But the big bright star is that our inventories should come back to normal in all markets by the middle of this year, which means that we will start seeing the demand as it happens.
Our next question comes from the line of Ross Seymore with Deutsche Bank.
Just want to ask a high-level one first. What do you think the normalized level of demand is for the company, whether you want to talk on a quarterly basis or annual basis? And I know that there's a ton of assumptions underneath that, and it's not the easiest to answer, but any shot would be appreciated.
Okay. I think we have answered this question on the last call. Our trend analysis would show that if everything came back to normal based on our share gains and market -- long-term market trend growth and so on. We should be really running at $150 million a quarter. However, it is unlikely, I believe, given the demand situation still being very weak, that it will happen in the second half of this year. It will probably be sometime again, I don't know what the '25 is going to be, but the earliest I can think of that we're getting there will be in the second half of 2025. And that's just a speculation on my part. It really depends upon demand and inventory levels all coming back to normalcy.
Got it. And then I guess, a gross margin question, kind of slightly nearer term and then a little bit longer term. Sandeep, you do a good job of talking about trends through the year. I know you said the second quarter is going to be up. So I guess the part A of the second question is utilization going up in the June quarter, seems a little odd if you just look at. You have basically a year of inventory on your books internally. And I know days in the channel is somewhat misleading at the bottom of the cycle, but how do we think about utilization? Why are you increasing it, given so much inventory on your books? And what's the trend through the year that you think you're going to have on gross margin if the back half revenue do indeed improve?
Yes. So first of all, we keep bulk of our inventory in wafer form. And when I talk about utilization, we -- yes, we do have some of our equipment are the fab power, but that's limited. But the bulk of the equipment is on the back end, which is the testers, handlers more machines. So the converting of the wafer into finished goods leads into higher utilization, the test outs. And that's what helps us in the margin.
But I think the bigger piece that is going to help us is going to be the yen. The yen has really moved quite a bit up in our favor with the dollar strengthening. But the funny part is that if you go a year back, it was in the 140s, then it dropped down to the 130 and then had gone back into the high 140s. So in Q1, as you turn the inventory, it's getting impacted by when it went down, but starting in Q2 and Q3, the favorability starts coming back and the yen will contribute quite a bit there.
Now we are also getting a mix benefit. As we have talked about before, we are going to see more consumer and industrial as the year goes by and less communication. For the year, I think I've talked about before, we should be somewhere in the 53.5%, 54%. That's the modeling I can do. With the pluses is coming from the yen mix in volume and the minus is coming from input costs. The input costs are still going up wafer and other costs. But as I said, even though I'm doing this high-level modeling, the mix is always the wildcard.
Our next question comes from the line of Tore Svanberg with Stifel.
Your line is open, please go ahead. Mr. Tore Svanberg. Our next question comes from the line of Matt Ramsay with TD Cowen.
I wanted -- obviously, we're working our way through the inventory correction across a number of markets and your company is going through it, no different than a lot of others. And we had a reset to the model on this call that we did 3 months ago, and you guys kind of walk through how you plan to progress through getting the inventory down, and what the model might look like.
I guess my biggest picture question is, as we've gone through the last 90 days, what's really changed other than you've progressed and started working the inventory down, and we're closer to coming out at the back end of it, but any big differences that in the last, say, 90 days as to what you expect it to play out? I mean it's not fun, but it's tangible progress. So I'm just trying to think if anything surprised you in the last 90 days or so?
Actually not. It is very much what we expected. We had said Q1 will be flat to potentially slightly up. But we really want to get the inventory down as much as possible. So we are happy that the inventory will come down again in Q1 based on our shipments, which is still below the demand, which really puts us in a good place for going forward from Q2 onwards. And the only other positive thing I would say is the consumers are really coming back. The consumer market is really coming back. We can see the orders being placed by people, who really stopped ordering for almost a year, a little more than a year now. They completely stopped ordering, and they are now coming back that tells us that they are out of inventory as well. It's not just our channel inventory, our end market inventory is also cleared up in consumer.
The consumer is in very good shape. Computer is also very close to normal in terms of inventory, it's the industrial that has extra inventory. And we believe that should come down to normal sometime in the second quarter. And we should start seeing bookings from industrial market is sometime in the second quarter. I don't know exactly when. So it's not that different from what we anticipated in Q4. And I think for the whole year, we still expect Q2 growth quarter and second half to be even better in terms of growth. And we think we could exit the year with a strong year-over-year growth in Q4 because of the comparison to the last year.
Matt, the other point to add is that nothing has changed in terms of what we were thinking for the year. The good part is it's playing out as we wanted, plus other people like if you see what Whirlpool said about the second half of 2024. Obviously, if they're going to see second half, we should see better a little earlier because of power supplies being made earlier. So it's good to see the validation of what we've been saying, and how it is playing out even from an outside party.
Got it. No, thank you very much both for all the detail there. That was my initial read as well, it's really steady progress, but not a ton of things have changed, which good in this kind of environment. I guess as my second question is a little bit more specific. And in the auto market. I think there was some commentary in the script about the pipeline of design opportunities in the EV market being considerably up maybe 80% or something like that. If you guys could maybe expand on that a little bit, both the nature of the opportunities and also if you win them, what the time to revenue could look like?
So I mean, to be honest, we are somewhat positively surprised by the level of interest we are getting on our products. I don't think it would have expected this a couple of years ago. If you had asked us, we would have expected a much slower ramp of design activity. So it really bodes well for the long term. And I'm getting more and more comfortable that this could be a $100 million business within the next 5 years. It has a long design cycle. But as we mentioned, we are already in production cars today at 8 different models in the market -- with 8 different customers, I should say. And that's quite surprising how quickly they adopted our products. Usually, the design cycles are very long.
Many customers are cutting their design cycle short to use our product simply because of the benefits we bring, whether it's size benefit, the component benefit, the component count benefit, the liability benefit and so on. And the other thing that's really surprising to me is even OEMs, who have historically been the hardest to address, like, for example, the Japanese OEMs are much more open because of the value we bring to this market. And that's true with not only Japanese, but also with the European customers and of course, the Chinese customers. So everything looks really good for automotive, and that's probably the most exciting growth area we are looking at right now.
Our next question comes from the line of Tore Svanberg with Stifel.
Yes. Can you hear me?
Yes, we can hear you, Tore.
Okay. Great. I don't know what happened there earlier. I could hear you guys, but you couldn't hear me. So my first question for you, Balu. So all downturns are different, that's pretty clear. But most people remind us that the upturns are almost always the same, meaning customers start to ramp pretty quickly. So anything that you're seeing this year that would be different? I mean I -- given your inventory comment, I assume that you sort of ready for a stronger ramp as and when it comes.
Yes, that's a good question. I wish I knew the answer. It's -- I just don't have a good feeling for the ramp of the recovery. I know it's going to recover. We think second half will be really good, but we don't know by how much. I wish I could really give you that answer.
Now let me talk about some areas where we are really feeling good. We already talked about automotive, but that's more longer term. But if you look at high power, the renewables are doing extremely well. And so we expect, again, another growth year on high power. The HBA market, which was very weak last year, we expect that to come back strong this year. Electric meters are doing extremely well -- electronic metros, I guess, are doing extremely well this year.
So there are several pockets in industrial that's going to come back. So I'm feeling really good that industrial will come back this year even though they still have some inventory. But when they come back, we will see a significant growth there. Consumer we already talked about. I think there inventory is clean, we're going to start seeing the demand. But the demand to the extent we can measure is still weak. So the growth in consumer in the near term is going to come from clearing up of inventory, not necessarily because the demand is increasing. We don't see that yet. Now that doesn't mean in the second half it won't come back strong. We just don't know about that. But just the inventory alone, clearing up will help us grow nicely in that market.
Of course, the -- we talked about cell phones before that is a declining market. We don't expect that to grow over the next few years. All of the growth is going to come from the other 3 markets, the computer, the consumer and industrial.
Very good. And on GaN, you highlighted several segments, several applications. The one segment that GaN could do well in, and you mentioned is data center, but we haven't quite heard enough from POWI as far as the traction you're getting there. So are you being conservative on when you penetrate that market? Do you already have traction there? Any update on GaN for data center would be great.
Good question. First of all, if you look at our SAM in 2027, if you go back to our Analyst Day presentation, we presented that we would double the SAM to $8 billion in 2027. Roughly about $3 billion of it will began SAM in that year. Of course, the GaN SAM will continue to increase. And out of that $3 billion, half of it we already have products for. We have products already in the market to address $1.5 billion of it. The other $1.5 billion comes from data centers, automotive, DC-to-DC converters and onboard chargers, telecom infrastructure like base station, power supplies and so on.
For that, we are working on products that are not yet ready, and until it's ready, we don't engage with customers. We have to get those products out. Now you may be wondering how come it takes us longer to get to market than some of the discrete guys. Well, if you have a discrete transistor you can go broadly in any market you want. But the problem with the discrete model is that then it becomes a commodity. You have a PIN compatible device that can be replaced by anybody, especially the Chinese guys, I think, are going to be incredibly aggressive in pricing of the discrete.
We just don't see us in the discrete market. It's just not the right business model for us. We have never been in the discrete market. And the way we make -- the margins we make compared to discrete guys, is by having a system-level solution that brings significant advantages over discrete. So that means it takes longer for us to come up with the products. It's the innovation of packaging, innovation of control schemes. There's a lot of things that we have to cover when we do a system-level product. So it takes longer. That's the bad news. The good news is when we come out, we have a very compelling product that can make very good margins, which is really hard to make with discrete. So that's why it takes longer. It's not that anything has changed. We did expect it to take some time. That's why we don't add that $1.5 billion of SAM until 2027.
Yes. No, that's helpful. Just 1 last one, Balu, on the automotive you talked about the design wins and so on and so forth. But from a revenue and timing perspective, could this be sort of like tens of millions in '25, or is that still more of a '26 time line?
That's -- yes, I would think by 2025, you should be -- let's see, let me -- it could easily be -- it will be, I would say, more than $10 million. I don't know exactly how much it will be. I'm a lot more comfortable saying, within 5 years, we would cross $100 million.
I would now like to turn the call over to Joe Shiffler for closing remarks.
Okay. Thanks, everyone, for joining this afternoon. There will be a replay of this call available on our website, investors.power.com. Thanks again, and good afternoon.
This concludes today's call. You may now disconnect.