TSMC (TSM) Q4 2014 Results Dicscussion

[user_2013101]

From the TSMC investor conference on Jan 15:

No direct comments on its top 2 or 3 customers.

2015 revenue will grow 16%. This forecast pretty much refutes a lot of rumors. TSMC cannot achieve such growth, when A8 production declines, 80-100% of A9 goes away, QCOM/AMD move to Samsung-GF, etc.

The only comment that may point to problems is that TSMC’s 14/16nm market share this year will be smaller than a big competitor, but regain commanding shares next year.

16nm volume production is Q2, no later than early Q3.

10nm tape-out and risk production by the end of this year.

This year’s capex $11.5 - $12 billion.

 

[Daniel Nenni]

Hsinchu, Taiwan, R.O.C., January 15, 2015 -- TSMC today announced consolidated revenue of NT$222.52 billion, net income of NT$79.99 billion, and diluted earnings per share of NT$3.08 (US$0.50 per ADR unit) for the fourth quarter ended December 31, 2014.

Year-over-year, fourth quarter revenue increased 52.6% while net income and diluted EPS both increased 78.5%. Compared to third quarter 2014, fourth quarter results represent a 6.4% increase in revenue, and a 4.8% increase in net income. All figures were prepared in accordance with TIFRS on a consolidated basis.

In US dollars, fourth quarter revenue increased 3.7% from the previous quarter and increased 46.4% year-over-year.
Gross margin for the quarter was 49.7%, operating margin was 39.6%, and net profit margin was 35.9%.

Shipments of 20-nanometer process technology accounted for 21% of total wafer revenues. 28-nanometer accounted for 30% of total wafer revenues. Advanced technologies, defined as 28-nanometer and 20-nanometer technologies, accounted for 51% of total wafer revenues.

“The strong demand for TSMC’s advanced technologies continued into the fourth quarter. Our 20-nanometer production was ramped-up at record speed and reached 21 percent of our fourth quarter revenue,” said Lora Ho, SVP and Chief Financial Officer of TSMC. “Despite a slightly weaker demand due to seasonality, we anticipate a more favorable exchange rate in the first quarter which will moderate the impact from seasonality. Based on our current business outlook and exchange rate assumption of 1 US dollar to 31.80 NT dollars, management expects overall performance for first quarter 2015 to be as follows”:

• Revenue is expected to be between NT$221 billion and NT$224 billion;
• Gross profit margin is expected to be between 48.5% and 50.5%;
• Operating profit margin is expected to be between 38.5% and 40.5%.

TSMC further expects the capital expenditures for 2015 to be between US$11.5 billion and US$12 billion.

About TSMC
TSMC is the world’s largest dedicated semiconductor foundry, providing the industry’s leading process technology and the foundry’s largest portfolio of process-proven libraries, IPs, design tools and reference flows. The Company’s owned capacity in 2014 was about 8.2 million (12-inch equivalent) wafers, including capacity from three advanced 12-inch GIGAFAB™ facilities, four eight-inch fabs, one six-inch fab, as well as TSMC’s wholly owned subsidiaries, WaferTech and TSMC China. TSMC is the first foundry to provide both 20nm and 16nm production capabilities. Its corporate headquarters are in Hsinchu, Taiwan. For more information about TSMC please visit http://www.tsmc.com.

 

[astilo]

Thank you Daniel and user_2013101.
Well, impressive 2014 performance as expected.
For the 2015 they are pretty much confirming the Q42014 numbers (no improvement). Considering the market is still growing and the favorable exchange rate, this forecast is good, but not excellent to me.

View attachment 12955

 

[Arthur Hanson]

The key figure is the increase in cap ex from an already high level. This is TSM's tell. There are many things they can't say, but cap ex tells all, especially if you listened to Lora Ho and Morris Chang over the years. They would in no way increase cap ex, unless they had a specific and profitable use to put it to. Apple and Nvidia both have game changing products coming online. Count on their high end fabs running flat out. They also have a relationship with Wuxipharmatec in what will become the largest market for semis. A very long story that's just starting. This could even dwarf customers like Apple in the future and not that far off.

 

[Fred P]

Bigger Capex = EUV?

 

[mliu1024]

No. EUV not ready. First phase of 10nm will use NXP3

 

[Arthur Hanson]

WX it seems has chosen the TSM platform for their medical mems and semis. WX currently has 8000 researchers and 3M ft2 of facilities in China and the US. They have even built manufacturing space in the US. They have a close relationship with the Mao Clinic. It seems TSM is in on the ground floor with the worlds first diversified bio foundry.

 

[Arthur Hanson]

In the interest of full disclosure I have a substantial long term holding in WX and TSM of over 10K shares in each. I have read over 20k pages on each.

 

[hist78]

Thank you Daniel and user_2013101.
Well, impressive 2014 performance as expected.
For the 2015 they are pretty much confirming the Q42014 numbers (no improvement). Considering the market is still growing and the favorable exchange rate, this forecast is good, but not excellent to me.

View attachment 12955

I assume you're talking about the 1Q2015 revenue forecast that is in the same range of 4Q2014 revenue. But this is still an impressive number because it's a 50% increase compare to 1Q2014.

 

[lefty]

But why did TSMC sell their ASML shares? Anyone know?

 

[hist78]

Some impression I got after watching the conference call:

1. TSMC took in NT$30 billion guarantee deposit from their customers for securing their shares in using TSMC's capacity. Last time TSMC took this kind of guarantee deposit is back to late 1990's. It's very interesting in terms market competition, technology strength, and capacity availability.

2. Dr. Morris Change stressed that TSMC builds capacity in response to customers' demand and not based on speculation. That means their customers are real and their revenue forecast is more close to real too.

3. After listening the conference call, I feel (purely by me) that TSMC won't sacrifice their operation flexibility, revenue, and profit margin just because one or two large customers' harsh demand. On the other hand, they are diversifying their customer base and go beyond Apple and Qualcomm while maintaining 2015 revenue growth several points above the 12% industry growth forecast.
.

 

[hist78]

But why did TSMC sell their ASML shares? Anyone know?

TSMC's CFO Mrs. Ho mentioned a word "hedging" during the conversation. So I guess in addition to all technology and market considerations, one of the reason to sell ASML shares well under the current market price is possiblely due to the "hedging" contract they made with an unknown third party has to be executed now. Not the best result in term the amount of money, but it's not a bad one either.

 

[Daniel Nenni]

Analyst reviews are coming in:

TSMC guides 1Q better than expected at flat q/q vs expectations for down -9-10%q/q despite fessing up to share loss on FinFET (Samsung 14nm is ahead of TSMC 16nm) -- better guide supports idea that AAPL C1Q is tracking better than seasonal on iPhone 6 demand, though investors are already nervous about C2Q weakness as AAPL orders decline and QCOM switches more business to Samsung due to Snapdragon 810 overheat issues. TSMC also raised capex guidance to $11.5-12B vs expectations around $10B and 2014 at $9.5B (15-20% higher than our forecast), with plans to add more 28nm capacity given strong demand for low-end 4G smartphones. TSMC guided for 20% of sales from 20nm in 2015 vs 10% in 2014. TSMC did not comment AAPL share at 16nm FinFET but said guidance for a high single-digit %age of sales to come from the 16nm in 4Q15 is intact, though TSMC expects its market share for 16nm FinFET to be much larger than those of its competitors in 2016, which jibes with the KGI call on foundry share. TSMC also plans to sell down its 5% stake (21M shares) of ASML. KGI Securities in Thailand

 

[user_2013101]

2. Dr. Morris Change stressed that TSMC builds capacity in response to customers' demand and not based on speculation. That means their customers are real and their revenue forecast is more close to real too.

I think, more precisely, TSMC builds capacity only with orders in hand, unlike its competitors building on speculations. This is the essence of Morris Chang’s comment.

How will TSMC spend the $12 B capex? 10nm risk production is intrinsically small in scale.

 

[user_2013101]

TSM stock price up 8% today, at 12:14 PM EST.

 

[user_2013101]

I think, more precisely, TSMC builds capacity only with orders in hand, unlike its competitors building on speculations. This is the essence of Morris Chang’s comment.

How will TSMC spend the $12 B capex? 10nm risk production is intrinsically small in scale.

20 nm revenue is forecast to be 20% for the entire 2015, similar to the 21% of 4Q14. Little increase or decrease at 20 nm. Perhaps expand 28 nm capacity somewhat; and upgrade older nodes for IoT and bio sensors. And, small scale 10 nm risk production.

A reasonable guess is that most of $12 B will be spent on 16 nm. And, TSMC does not build capacity unless orders in hand.

However, the contradicting points are two: 1) 16nm market shares smaller than a competitor; 2) 16 nm revenue is only high single digits at 4Q this year.

Either TSMC overstated the capex, or underestimated its 16nm revenue (perhaps to lowball the expectation for a upward surprise?).

 

[lefty]

20 nm revenue is forecast to be 20% for the entire 2015, similar to the 21% of 4Q14. Little increase or decrease at 20 nm. Perhaps expand 28 nm capacity somewhat; and upgrade older nodes for IoT and bio sensors. And, small scale 10 nm risk production.

A reasonable guess is that most of $12 B will be spent on 16 nm. And, TSMC does not build capacity unless orders in hand.

However, the contradicting points are two: 1) 16nm market shares smaller than a competitor; 2) 16 nm revenue is only high single digits at 4Q this year.

Either TSMC overstated the capex, or underestimated its 16nm revenue (perhaps to lowball the expectation for a upward surprise?).

Not all the CapEx is 16nm. They're probably expanding 28mn (Also ultra-low power and 28HPC). TSMC had to turn away customers last year because hadn't enough 28nm.

 

[Daniel Nenni]

The conference call transcript is out:

Taiwan Semiconductor's (TSM) CEO Mark Liu on Q4 2014 Results - Earnings Call Transcript | Seeking Alpha

It is definitely woth a read. There are some surprises!

 

[hist78]

TSMC's CFO Mrs. Ho mentioned a word "hedging" during the conversation. So I guess in addition to all technology and market considerations, one of the reason to sell ASML shares well under the current market price is possiblely due to the "hedging" contract they made with an unknown third party has to be executed now. Not the best result in term the amount of money, but it's not a bad one either.

From the conference transcript:

Taiwan Semiconductor's (TSM) CEO Mark Liu on Q4 2014 Results - Earnings Call Transcript | Seeking Alpha

Lora Ho - SVP and CFO

"My last comment is about ASML stock sale. As you know, in August 2012, we acquired about 21 million shares of ASML. Under its customer co-investment programs, the purchase price was €39.91 per share for a total of $838 million. There was a lock up period of 225 years. In the last two years TSMC has entered several hedging contracts, now fully covered our position with an average hedge price of €62.59 per shares, resulting in a locking profit of €483.5 million. As the lock up period is to be expired in April this year, we will be able to book a total profit of about NT$21 billion in 2015. This one-time non-op gain is expected to increase our EPS by NT$0.61 in second quarter and NT$0.13 in the third quarter and for the full year will be about NT$0.75."

IMHO, for TSMC to own ASML stocks is kind of conflict of interest. Now TSMC has more freedom to make decision without worrying about hurting TSMC's own ASML investment.

 

[Li Yisuo]

IMHO, for TSMC to own ASML stocks is kind of conflict of interest. Now TSMC has more freedom to make decision without worrying about hurting TSMC's own ASML investment.

CFO cares only about financing for the company. So buy cheap, hedge risk, and sell high. When interest rate starts to mount up, volatility will shoot up. So now it is the best time to reap the profit on your investment. Great work, Lora Ho.