KULICKE & SOFFA INDUSTRIES KLIC
April 14, 2015 - 4:25pm EST by
jet551
2015 2016
Price: 15.93 EPS 0 0
Shares Out. (in M): 77 P/E 0 0
Market Cap (in $M): 1,233 P/FCF 0 0
Net Debt (in $M): -517 EBIT 98 115
TEV ($): 716 TEV/EBIT 0 0

Sign up for free guest access to view investment idea with a 45 days delay.

  • Manufacturer
  • Duopoly
  • Cyclical
  • Activism
  • Buybacks
  • Semiconductor
  • Semi cap equipment

Description

Kulicke and Soffa (K&S) is the market-leading manufacturer of wire bonders, a key piece of equipment in the semiconductor packaging process.  It is a cash cow with underappreciated growth prospects that has close to ~45% of its market value ($7/share) in cash, and is only trading at 6.6x EV/TTM EBITDA (profroma recent acquisition).  Our forecast calls for $2.75 of cash flow generation over the next eight quarters, resulting in a cash balance of ~$9.25/share in cash or roughly 60% of its current market value, and a 28% increase in EBITDA (combination of organic growth and a small acquisition that closed in January).  We think K&S is worth at least ~$23/share, or ~45% upside from the $15.93 it trades at today. In addition to a large cash buffer, strong cash flow, and attractive growth prospects, there’s also a “free option” from their new line of thermo-compression bonders (TCB) currently in beta testing with potential clients.  We’re not currently modeling any sales for this new segment, but it has the potential to add significantly to profitability starting in 2016, and if it does, K&S will be worth much more than our $23/share target.

K&S has been written up on VIC before, and we refer you to those write-ups for additional background.

 

THE WIRE BONDING “CASH COW” BUSINESS

People have been saying wire bonding (ball, wedge and stud-bump bonding) is an outdated technology and should start to decline for years – ever since flip chip technologies were introduced decades ago.  However, while flip chip and other advanced packaging technologies have grown faster than wire bonding, wire bonders have continued to get better and cheaper, and wire bonding has continued to grow as well.  The installed base of wire bonders has grown from 80k machines to ~130k over the past 5 years, and wire bonding continues to be how the majority of integrated circuits are connected. This trend is expected to continue, and literally every one of the 25+ industry consultants, wire bonding experts, wire bonder manufacturer employees, and wire bonder purchaser employees we spoke to said they expect wire bonder sales to continue to grow in the low – mid single digits going forward.  One analyst at a leading consulting firm said it well; “Wire bonding is the freaking energizer bunny, it just keeps going and going, and won’t be stopping any time soon”. 

However K&S is priced like wire bonding is a declining market, not the increasingly stable and growing cash cow market that it is.  Based on our research, here are our forecasts for K&S’s WB sales over the next few years.

 

 

Wire bonding has and will continue to grow because packaging is primarily about cost, and wire bonding continues to be the lowest cost packaging option for the vast majority of package types.  It’s also highly flexible, and can be used on a variety of IC packages where flip chip is just not possible or is too expensive.  So while flip chip and other technologies are gaining market share mainly in use cases that require small packaging, wire bonding is still used for the bulk of interconnections.  The picture below illustrates this – even in a high-end smartphone wire bonders are still used to connect a number of the components.  

 
 

The emerging Internet of Things market will likely heavily utilize wire bonding as well.  K&S dominates the wire bonder market with ~65% share, and is in a stable duopoly with ASM, which controls ~20-30%.  K&S is the clear technology leader, while ASM is the low cost option.  Our conversations with both companies and other experts in the field imply that this dynamic is stable.  A senior manager at K&S said that they haven’t seen a highly competitive dynamic with ASM in quite a while, nor do they expect any going forward. This was corroborated by ASM’s IR director who said that they weren’t making significant investments in wirebonding vs. advanced packaging, where they are currently focused.  One of the semi-equipment sell side analysts we spoke to said that ASM tried to gain share by competing aggressively on price in the late 90s, but it didn’t really work and pricing dynamics have been stable for years.  Another senior manager of a niche wire bonding competitor who’s been in the industry for 30+ years said that he isn’t seeing any competitive pressures and his impression is that both K&S and ASM are happy with the positions they occupy in the market.

Wire bonding is a classic “cash cow” for K&S that we think investors are overlooking because it isn’t a sexy new technology.   

Additional positives for K&S’s wire bonding business:

  • K&S has stable customer relationships from the diligence we’ve done.  A senior VP at their largest client, ASE (usually 10-20% of sales) said: “K&S provides 100% of our wire bonders, and our relationship with them has enabled us to become the world leader in copper wire bonding.  I don't see our relationship changing going forward.”
  • K&S dominates the copper wire bonder market, with 90% of copper wire bonder sales.  The switch to copper drove large increases in sales from 2010 – 2012.  While those results aren’t likely to be repeated, the transition to copper still has legs as ~50% of bonders are copper focused, and eventually the market will be 70 – 75% copper.  K&S forecasts imply a similar transition pace in 2015/2016 as in 2013/2014 of ~5% increase in share. 
  •  After the transition to copper slows down, the replacement market will start to increase dramatically.  Wire bonders are typically replaced every 6 – 10 years depending on how leading edge the operator is (K&S’s customers are generally the most advanced).  With large additions to the installed base occurring in 2010 – 2012, it’s likely that there will be a corresponding sales spike somewhere in the 2016 – 2020 timeframe.

 

THE THERMO-COMPRESSION (TCB) OPPORTUNITY

TCB is a new advanced packaging technology that enables even smaller interconnections (finer pitch) than standard (mass reflow) flip chip technologies.  It is currently in its very early stages (~100 machines in use now), but should be one of the next big areas of growth in semiconductor packaging over the next decade, with management expecting the market to reach ~$400mm in annual sales by 2016, and $600mm in sales by 2017. K&S has two internally developed TCB offerings (chip-to-substrate and chip-to-wafer), and is initially targeting the 3D stacked memory market.  Their bonders are still in beta testing, but we know that all of the big memory players are currently qualifying K&S’s products, and that the feedback has been positive.  In addition to creating finer pitch bonds than standard flip chip, TCB is also superior because using it in combination with another technology called through-silicon-via (TSV) to stack dies results in a significant performance increase over traditional packaging as bandwidth increases while power consumption decreases.  Major memory suppliers (Micron, SK Hynix, and Samsung) haven’t made the jump to 3D stacking yet, but clearly will due to the performance benefits, and once one does they all will have to, in order to stay competitive.  An informed party at a major Integrated Device Manufacturer (IDM) explained, “we don’t know when TC Bonding will take off, but when it does it will explode”.

Management hasn’t given very specific guidance yet, but has suggested $100mm in sales as a reasonable target once the market scales up.  They expect to see some sales in 2015, but 2016 will be the year that sales really start to ramp up.  We cross-checked management’s assumptions with other major players and potential customers in the space, and everyone we spoke to thought those numbers were possible.  However, visibility into timing/future market share is low, and 2016 could be the year TCB takes off, but it’s also possible that sales don’t reach significant volume until 2017.  An SVP at Amkor (a large K&S customer) told us that management’s $400mm market forecast was probably optimistic for 2016, and that it could take longer to hit those numbers.  Two leading industry consultants said something similar, indicating it is reasonable-to-optimistic to assume 2016 is the year the market reaches the ~$400mm number, but that it should happen at some point in the next few years.  All said that K&S is a leading contender to get a big slice of the market.

The current competitors in the space are ASM Pacific, BE Semi, and Toray.  K&S is a later entrant to the market, but has invested significantly in R&D (~$35mm/year in 2014, $25mm expected in 2015).  We’re told their product offerings have caught up to competitors and will be ahead within the next 6 months.  We’re not experts in the space so it’s hard for us to opine directly on this advanced technology, however the people we’ve spoken to have said this won’t be a winner take all market, and that at least 3 of the 4 players are likely to take some share, likely in a 60/20/20 split.  So even if K&S does not end up being the dominant player, it’s still very likely that they can get to something like $80mm in sales from TCBs sometime in the next few years.

We’re not modeling any sales from TCB, AND still including $25mm/year of R&D costs when we arrive at our $23 price target.  So TCB is a free option in our thesis, and a pretty big one at that.  We’re modeling it this way because we’re not experts on the technology and don’t have visibility into timing.  However, even in a downside scenario TCB is still likely to provide $80mm in revenue/$20mm in EBITDA per year.  And in a complete disaster scenario where the program fails you’d still likely get $25mm in R&D cost savings.  So we think we’re being conservative here, especially since the upside is huge – this will be an incredibly fast growing market and K&S is very well positioned with a great track record of execution (see copper wire bonding).  Our contacts at K&S and Amkor have confirmed that we should have a much better sense for how the TCB market will shake out by the end of this year, giving K&S a possible catalyst in that timeframe. 

 

HISTORICAL VOLATILITY

K&S has had a good amount of volatility in their historical sales, both due to industry volatility and company specific factors. However recent developments in both the semiconductor industry and the company suggest that volatility could be somewhat reduced going forward.

Semiconductor Cyclicality

We are not semiconductor experts and have traditionally avoided the sector because of the technology complexity and its reputation for dramatic cyclicality.  However, in researching K&S, a recurring theme emerged that absent a large shock to the global economy, the volatility of the traditional semiconductor cycle should be reduced going forward.  For instance, Bill McClean, President of IC insights, said in his keynote address at the 2015 Device Packaging Conference “I won't say cycles are dead, but we're only going to see a big downturn if we have a large departure from trend in the global economy”.  A senior employee at ASE agreed, saying that historically the major cycles were driven by “stupid” overinvestment cycles, and based on his experience in the industry it’s highly unlikely we see that level of stupidity again given past lessons and consolidation across the industry.  Semi focused buy side analysts agreed as well, saying they “hate to say it’s different this time”, but it certainly looks like cyclical volatility will be lower going forward. 

We’re generally pretty skeptical of statements like this, but the logic makes sense to us.  Historically, volatile cycles have come from large over investment, which led to over capacity, falling prices, and then declines.  Many of the larger cycles were driven by uneconomic nationalistic behavior (Korea, Taiwan, etc. trying to create a national industry), or cutthroat competition in an emerging field.  Dynamics appear to have improved; the industry is simply much larger and more diverse than it used to be, it’s much more driven by consumer demand than corporate demand (60/40 vs. 40/60 a decade ago), it’s grown more consolidated, and the major players have survived through a numbers of cycles and have hopefully learned from past errors.  

Company Specific Volatility

In addition to reduced industry volatility, K&S’s product mix should be less volatile going forward.  The company is growing increasingly diversified as non-wirebonding areas grow faster than their core business, including the recent acquisition of Assembleon, which is about 15% of pro-forma revenue.  This should smooth out seasonal volatility somewhat and make K&S a bit more immune to wire bonding specific disruptions.  In addition to these product mix shifts, wire bonding orders themselves will also likely be more stable going forward.  Sales from 2010 – 2014 were lumpy due to a huge shift to copper wire bonders in 2010 – 2012 from the largest OSAT companies, and then a big decline in orders from those same companies in 2013 -2014.  The continued transition to copper wire bonders should be more stable over the next few years, and orders from those large OSATs should normalize.  These changes won’t mean K&S is no longer a cyclical business, but it should certainly be more stable than it has been historically.

 

ASSEMBLEON

While the Assembleon acquisition seemed somewhat expensive to us (10x fwd EBITDA), the acquisition makes sense strategically, and could provide significant upside.  Assembleon has a line of equipment focused on the surface mount technology (SMT) space, a traditional area of packaging that occurs later in the process than wire bonding, and also has flip chip and wafer level packaging machines which are in the high-growth advanced packaging segment.  Both of these business lines have overlap with K&S’s current customer base.  

Assembleon is around 15% of overall revenue/EBITDA, and based on our conversations we think that management is being conservative with their forecasts for the business.  A contact we spoke with said that he expects Assembleon sales growth to be more like 20% vs. the 10% that management has publically stated.  He also thought it was possible they sell $100mm of flip-chip and other advanced packaging equipment within a year or two, from ~$60mm last year.  We’re modeling ~15% growth for Assembleon over the next two years, higher than management’s guidance, reflecting this due diligence.  

 

CASH USES

Pro-forma for the Assembleon acquisition K&S had ~$535mm of cash on their balance sheet (~$7/share) at the end of the year relative to a market cap of $1,200mm.  This cash gives them an enormous cushion to weather any potential downturn.  One potential worry stemming from this excess cash is that it will be used for a stupid acquisition, however we think that risk is low.

  • K&S has been trying to do an acquisition since 2010 and looked at dozens of companies before acquiring Assembleon, suggesting that they are relatively disciplined.
  • Management has told us it is unlikely they’ll make another acquisition in the 12 months following the Assembleon acquisition.
  • We’ve been told by a contact that K&S doesn’t think they need to acquire another equipment company given their current portfolio of advanced packaging products.  If they do make an acquisition it will be in the metrology/data analysis space, which would be a complementary service to existing offerings.  Our understanding is that there are a number of small companies in these areas ($25 – 100mm), so we think it’s less likely they’d do a large ($200mm+) acquisition. 

The other obvious use of their cash is returning it to shareholders.  On that front they have already committed to pursuing a $100mm buyback.  We believe that the large amount of cash still on the balance sheet is partially a hedge to make sure that they still have flexibility if their internal advanced packaging lines don’t work out.  So if they get visibility that advanced TCB is taking off in the next 6 - 12 months we think returning more cash to investors in highly likely.  Another point worth making is that there is now an activist investor (Lemelson Capital) who has told us he intends to push aggressively if management doesn’t continue to return cash in an intelligent way. 

 

FORECASTS AND VALUATION

K&S currently trades at 6.5x, which we think is very cheap given their market leading position in wire bonding, exposure to higher growth advanced packaging markets (TCB and Assembleon, which they just acquired), and impeccable balance sheet.  Here’s how we see the business growth over the next two years:
 
 
We use an 8x multiple for K&S given trading comps that average around 14x range (9x is the lowest), and the fact that K&S’s fundamentals are just as strong as their peer group.  K&S appears to be trading at a discount because 1) People have misconceptions about the prospects of wire bonding 2) it’s not particularly well followed given lackluster sell side research, value characteristics in a cyclical/growth industry, and the fact that it’s headquartered in Singapore but listed in the US, and 3) investors aren’t giving them full credit for cash on their balance sheet, which we think is unjustified given the variety of ways they can utilize that cash (while avoiding repatriation taxes), and their demonstrated commitment to returning cash to shareholders.
 
 

 

K&S is also cheap by historical standards.  Historically profits have been very volatile so it’s helpful to look at valuation based on a trailing average EBITDA (in this case we’re using 7 years).  By this measure K&S has traded around 8.5x on average vs. the current 6.8x (7yr multiple).  

 
 
 
 
I do not hold a position with the issuer such as employment, directorship, or consultancy.
I and/or others I advise do not hold a material investment in the issuer's securities.

Catalyst

  • In 2013 and 2014, K&S closed their factory for the week of Chinese New Year.  However, we’ve learned that the factory remained open this year, which could mean good Q2 performance if orders remained strong for the rest of the quarter.  Results will be announced in early May.
  • Positive news continues to come out this year on the new line of TC bonders
  • Continued strong cash generation
  • Management continues to buy back shares
    sort by    

    Description

    Kulicke and Soffa (K&S) is the market-leading manufacturer of wire bonders, a key piece of equipment in the semiconductor packaging process.  It is a cash cow with underappreciated growth prospects that has close to ~45% of its market value ($7/share) in cash, and is only trading at 6.6x EV/TTM EBITDA (profroma recent acquisition).  Our forecast calls for $2.75 of cash flow generation over the next eight quarters, resulting in a cash balance of ~$9.25/share in cash or roughly 60% of its current market value, and a 28% increase in EBITDA (combination of organic growth and a small acquisition that closed in January).  We think K&S is worth at least ~$23/share, or ~45% upside from the $15.93 it trades at today. In addition to a large cash buffer, strong cash flow, and attractive growth prospects, there’s also a “free option” from their new line of thermo-compression bonders (TCB) currently in beta testing with potential clients.  We’re not currently modeling any sales for this new segment, but it has the potential to add significantly to profitability starting in 2016, and if it does, K&S will be worth much more than our $23/share target.

    K&S has been written up on VIC before, and we refer you to those write-ups for additional background.

     

    THE WIRE BONDING “CASH COW” BUSINESS

    People have been saying wire bonding (ball, wedge and stud-bump bonding) is an outdated technology and should start to decline for years – ever since flip chip technologies were introduced decades ago.  However, while flip chip and other advanced packaging technologies have grown faster than wire bonding, wire bonders have continued to get better and cheaper, and wire bonding has continued to grow as well.  The installed base of wire bonders has grown from 80k machines to ~130k over the past 5 years, and wire bonding continues to be how the majority of integrated circuits are connected. This trend is expected to continue, and literally every one of the 25+ industry consultants, wire bonding experts, wire bonder manufacturer employees, and wire bonder purchaser employees we spoke to said they expect wire bonder sales to continue to grow in the low – mid single digits going forward.  One analyst at a leading consulting firm said it well; “Wire bonding is the freaking energizer bunny, it just keeps going and going, and won’t be stopping any time soon”. 

    However K&S is priced like wire bonding is a declining market, not the increasingly stable and growing cash cow market that it is.  Based on our research, here are our forecasts for K&S’s WB sales over the next few years.

     

     

    Wire bonding has and will continue to grow because packaging is primarily about cost, and wire bonding continues to be the lowest cost packaging option for the vast majority of package types.  It’s also highly flexible, and can be used on a variety of IC packages where flip chip is just not possible or is too expensive.  So while flip chip and other technologies are gaining market share mainly in use cases that require small packaging, wire bonding is still used for the bulk of interconnections.  The picture below illustrates this – even in a high-end smartphone wire bonders are still used to connect a number of the components.