Cohu had begun reporting a string of positive earnings several quarters before their acquisition of Xcerra in October of 2018.
This was a small debt free company that had several acquisitions that had essentially rolled up the several type of test handlers made globally.
Their first start was when the began building heat chambers for a small company called INTEL.
My first shares purchased were in 1978.
They were debt free all those years until the purchase of Xcerra in 2018.
Previous acquisitions (other than small technology buys) were Rasco - the then number 2 gravity feed handler. This was a purchase in December of 2008 - right during the big collapse. They bought Rasco from Dover (who also owned the number 1 maker of gravity feed handlers Microtest). The Chairman of the Board, James Donahue knew how to build parallel gravity test handlers as they had already done that with pick and place handlers (their expertise developed by adhering to INTEL'S specs). It was his intention to make the gravity feed handler of Rasco's faster then Microtest's and at a price that it would make a good margin. This was slowly accomplished in the next 4 to 5 years. and Rasco is now the number 1 market share holder of Gravity feed handlers
This was a big acquistion and took their cash from 155 million to 70 ish million. in 2008.
Once they had built cash back to 160 million, they bought Ismeca (a Swedish test handler that had come up through the sewing machine background). This provided a Turret test handler with the # 1 market share . The Swedish company sold off Ismeca to pursue making higher margin wind mill blades. Donahue stuck with what his expertise was.
By this time they had rolled up the global type of handlers and achieved a number one market share in each sector.
Their only other competitor was Xcerra, and they were a contract outsourcer for most of their handlers. Xcerra then bought from Dover Microtest, the now # 2 handler that had been number 1 when Cohu bought Rasco also from Dover.
Xcerra then marketed itself to a Chinese hedge fund,and an acquisition had been proposed. This required approval from the CFIUS:
The Committee on Foreign Investment in the United States ...
home.treasury.gov/policy-issues/international/...
The Committee on Foreign Investment in the United States (CFIUS) CFIUS is an interagency committee authorized to review certain transactions involving foreign investment in the United States and certain real estate transactions by foreign persons, in order to determine the effect of such transactions on the national security of the United States.
Cohu wrote a white paper on how China could control the cost of all chip testing if they were to be subsidized by the CNP. CFIUS agreed Cohu's view and did not approve the purchase.
That then put Cohu, a smaller company,with less cash than XCERRA (but still 160 million), and smaller margins that Xcerra had.
It was a big bite for Cohu compared to All of their previous acquisitions - which all were rightsized and rolled in nicely, but with a time lag of two years for each of the larger acquisitions.
Prior to the Xcerra acquisition Cohu had also bought a Japanese firm Kita. Kita made the spring pin point contactors that actually touch the chip during the test cycle. Kita also was from the sewing machine background - I'm positing they learned of them from the Ismeca acquisition.
During this time, Cohu was finding that upon rolling up all of the the handler types business on a global basis, They in fact had the largest installed owner base of handlers.
Customarily,when one sells a handler, their service force (which was world wide for years) were counted on to maintain the handlers and sell spare parts - especially the contactors that wear out. Cohu calls that business "consumables".
It's a very good business as margins are in the mid to high 60's and it is stable.
During this pre Xcerra acquisition Cohu had achieved a penetration level of consumables such that they were 48 to 52% of revenue and with no big changes in their business model a breakeven was obtainable, in a market where new Capex handlers was not happening - (cohu refers to this revenue stream as "System sales").
So they were in a fairly good shape to consider the very big next acquisition of Xcerra.
Keep in mind Xcerra had bought Microtest from Dover and three or four years befre that Microtest was the largest gravitiy handler with number 1 market share, but had lost it to Rasco with their now faste parallel gravity handler.
The acquisition of Microtest's established ownerbase (now owned by Xcerra) would really give Cohu hands down the scale on the consumables business.
That part made instant "YES let's do it" sense.
Xcerra had also ventured into actual contract testing in Taiwan and mainland China for Chinese chip makers.
Additionally they had:
a Printed Circuit board inspection business
a contactor business which mates the handler to the tester
a wafer inspection business
Facilities all over the globe as well.
So to absorb this global organization Cohu for the first time took on 355 million in debt and expanded their shares outstanding from 27 million to 42 million.
The debt was arranged for a 5 year basis.
During the last 2 years 5G was the big forward demand for "System sales". The ramp has been slow to develop, but now appears to be gaining traction. As 5G mobile handsets are now being offered, it appears we are just in the beginning phase of the retail handset ramp. This will be huge for Cohu as it ramps.
Xcerra also brought to the party a business that made pieces for the infrasytucture of 5g and that has been a bit of a boost.
That being said the ramp and boost of system sales haev been slow to come.
Just this month Cohu guided that their earnings would be on the high side of guidance. They are seeing system sales for the more RF chips built into the 5G handsets. It is a positive word that has been absent since October 2018.
I believe it is safe to say that Cohu, haf they not made the acquisition would have been modestly profitable over the last 2 years.
During the last earnings webcast Cohu suspended the dividend and is now paying down its debt from the 355 million level. They had been servicing interest only since.
Donahue is a very conservative Chairman of the board. His entire tenure at Cohu (from the 1980's) had never had debt. He has never increased the dividend in his tenure - he always preferred acquisitions and maintaining the dividend that was pad before he was CEO and Chairman.
To his credit he redefined the company several times, made small acquisitions and made them become accretive.
The slow to develop new business has been a burden on Cohu's profitability.
The extra shares did create a larger dividend pay ut, which seems to now be resolved.
Jeffrey jones is an excellent CEO. He was hired after GE bought out SBSE (I also owned this stock). He brought tax deferrals and non gaap acounting to a small growth company, which has left free cash flow to maximize.
Jeffrey's pro-forma cash revenue was at the 800 million level.
Reality has ranged form closer to 600 million.
So the expected revenue levels have been lacking and it is safe to say that system sales have been absent to the degree they expected.
The handler business is very cyclical, the delay of 5G ramp is and was what was looked forward to.
I believe we are now on the cusp of system sales ramping and it should do very well into 2021.
Cohu NEVER discusses orders and backlog. The most recent investor presentation they attended they mentioned strong order for mobility and RF chipsets - that's Cohu speak for 5G.
These guys are good and conservative.
They took a big plunge and bought XCERRA.
The cost of rightsizing the combined company has killed any earnings since.
The dividend has been suspended and they paid down the acquisition debt by 17.3 million this last quarter. I suspect Donahue is sleeping better , but pushing for revenue growth and system sales with margins in tact.
This company is a turnaround story of a cyclical industry that has now rolled up its global business and I believe achieved scale that will begin to show when the system sales kick in.
I believe that is happening as I type!
Now keep this statement in perspective. Cohu has amassed scale in consumables and this last quarter they actually represented 52% of their revenue.
This puts them breaking even and covering R&D.
If you look back in history at Amat (when they bought Varian), or Lam Research (when they merged with Novellus) they marked moments when their segment achieved mass and scale. Break even was achieved by consumables and margins were steady as they had consolidated the competitors under one roof.
They never talk about orders, but mentioned strong orders this last month in a preannouncement.
They have a virtual investor presentation with Deutsche Bank October 6th.
Earnings I'm guessing will be announced the 3rd or fourth week of October.
If guidance moves to 175 million per quarter it will put them into very nice profitability and well on schedule to achieve their mid term goal of 200 million per quarter and $3.00 a share non gaap EPS.
If they beat and guide strongly, I'd not be surprised to see the past high of $27.00 ish to be pushed by year end.
I've been saying this for years and even decades, but when an Amat or an Avago realize that Cohu has the ability to control the cost of every chip made in the world, there is going to be a hefty premium paid by someone who sees that as a very quick way to payback the purchase and assure the margins on your chip while making others pay a bit more for theirs.
That's my story, I can't prove it but I'm sticking to it! <Smile>
Sorry for the long post, but I do think it is a multi-bagger in the works. It represents a turnaround story and a great value amongst many that have already had their 4 baggers up.
Cohu's past high was in 2000 @ $61.75.
My bet it we're going to see it again!
Jeffrey jones indicates needed cash to continue Capex and R&D runs 50-70 million as I recall.
164 million is right at the most cash than they've ever had before and is a very solid level for them.
This one is going to pay for my boat! LOL
Bob
|
