台積電 ADR (TSM) 2009 Q3 法說會逐字稿

Welcome to TSMC's third quarter '09 results webcast conference call.

This conference call is being webcast live via the TSMC website at www.tsmc.com and only in audio mode.

Your dial-in lines are also in listen-only mode.

I would now like to turn the conference over to Dr.

Elizabeth Sun, TSMC's Head of Investor Relations.

Thank you, Dan.

Good morning and good evening, everyone.

Welcome to TSMC's third quarter 2009 conference call.

Joining us on the call are Dr.

Morris Chang, our Chairman and Chief Executive Officer, and Ms.

Lora Ho, our Vice President and Chief Financial Officer.

The format for today's conference call will be as follows.

First, Lora will summarize our operations in the third quarter and give you our guidance for the fourth quarter.

Then TSMC's Chairman, Dr.

Chang, will provide his general remarks on the business outlook and a couple of key messages.

Then we will open the floor to questions.

For those participants who do not yet have a copy of the press release, you may download it from TSMC's website at www.tsmc.com.

Please also download the summary slides in relation to today's quarterly review presentation.

I would like to remind all listeners that the following discussions may contain forward-looking statements, subject to significant risks and uncertainties, which could cause actual results to differ materially from those contained in the forward-looking statements.

Information as to those factors that could cause actual results to differ materially from TSMC's forward-looking statements may be found in TSMC's annual report on Form 20-F, filed with the United States Securities and Exchange Commission on April 17, 2009, and such other documents as TSMC may file with or submit to the SEC from time to time.

Except as required by law, we undertake no obligation to update any forward-looking statements, whether as a result of new information, future events or otherwise.

And now I would like to turn the call over to Lora.

Thank you, Elizabeth.

Good morning and good evening to everyone.

Welcome to our third quarter 2009 earnings conference call.

First, I will go over the highlights of our third quarter.

Then I will give you the outlook of the fourth quarter 2009.

Please refer to the quarterly financial summary slides on our website.

All dollar figures are in NT dollars unless otherwise stated.

Now, let me go over some highlights for the third quarter.

On page four, in the third quarter, our business sees continued improvement across all segments.

With increased utilization rate, we delivered improved margin and profit.

Compared with the second quarter, our profitability has recovered back to the level before the financial crisis.

Net sales were TWD89.9b, 21.2% increase over the last quarter, and declined only 3.3% year over year.

Third quarter wafer shipments was 2.45m 8-inch equivalent wafers, representing a 24% sequential growth and a 1.4% increase year over year.

Gross margin and operating margin were 47.7% and 35.6%, respectively.

Comparing with last quarter, it was an increase of 1.5 percentage points and 1.7 percentage points, respectively.

Comparing with year over quarter, gross margin and operating margin increased 1.4 and 0.2 percentage points, respectively.

EPS for third quarter '09 was TWD1.18.

Our ROE went up to 27.3% from prior quarter's 21.4%.

Let's take a look at our income statements.

Our gross margin rate was 47.7%, representing a 1.5 percentage point increase from last quarter.

The improvement was primarily due to increased utilization rate and continued cost improvement, partially offset by an unfavorable exchange rate.

Operating expense increased TWD1.7b from second quarter, reflecting increased research activities in our 28 and 22-nanometer technologies.

We also registered a higher legal fee and a one-time donation for typhoon relief program.

Non-operating income was TWD0.52b, lower than TWD1.15b in the second quarter '09.

The sequential decline was primarily due to the absence of litigation compensation and less interest income.

Net investment gain was TWD450m, higher than TWD106m of prior quarter, mainly due to continued business improvement among invested companies.

Net margin was 34%, representing a 1.1 percentage point increase, both sequentially and from year-ago quarter.

On page six, now let's examine our revenue by applications.

On a quarter-on-quarter basis, we saw wafer sales increase across the board, computer grow most with a 29% sequential increase, followed by communications, 25%, and consumers, 11%.

Overall, revenue from communications, computer and consumer applications accounted for 46%, 30% and 18%, respectively.

In terms of revenue by technology, total wafer sales from advanced technologies accounted for 67% of our total wafer sales, representing a 2 percentage point increase from last quarter.

Moreover, 40-nanometer and 65-nanometer combined accounted for 35% of our total wafer revenues.

For 40-nanometer, revenues were more than quadrupled, compared with the second quarter, accounting for 4% of our total wafer sales.

For 65-nanometer, revenue contribution was 31%, 3 percentage point increase from last quarter.

Meanwhile, 90-nanometer and 0.13-nanometer represented 18% and 14% of our total wafer sales, respectively.

Now let's move on to balance sheet and cash flow statements.

We ended the third quarter with TWD180b in cash and short-term investments, down from TWD247b in prior quarter, primarily due to the payment of cash dividends, which also led to the decrease in our current liabilities.

Accounts receivable days and inventory turnover days were 36 and 41 days, respectively.

Our net fixed asset turnover was 1.6 times.

In cash flow, total cash inflow generated by operating activities reached TWD47b, compared with TWD25b in last quarter.

The sequential increase was mainly due to increased business activities.

In terms of cash outflow, we paid TWD77b for cash dividends.

Capital expenditure was TWD32b.

Meanwhile, roughly TWD20b was used to purchase short-term marketable securities, mainly for fixed-income instruments.

In sum, our ending balance was TWD157b, down TWD83b from prior quarter.

Now, I'm in page 10.

Now let's turn to capacity and CapEx.

Total installed capacity was about 2.44m 8-inch equivalent wafers in the third quarter, a small decline from the prior quarter.

The decrease was due to capacity migration into advanced nodes.

For 2009, overall capacity is forecast to be around 9.96m 8-inch equivalent wafers, a 6% increase over 2008.

Advanced technology capacity is forecast to increase by 11% year over year, reaching 42% of total capacity in 2009.

For CapEx, we spent $971m during the third quarter and $1.4b year to date.

In order to meet the rapidly growing demand in 40 and 65-nanometer technologies, and in the end of this year and moving to 2010, we will be more aggressive in capacity investment in 12-inch advanced technology.

Now let me give you our outlook for the fourth quarter of 2009.

Based on current business expectations and a forecast exchange rate of TWD32.1, we expect our consolidated revenue in fourth quarter of '09 to come in between TWD90b and TWD92b.

In terms of margins, we expect fourth quarter gross margin to be between 47% and 48.5%, operating margin to be between 35.5% and 37%.

2009 capital expenditure will be increased to around $2.7b.

This concludes my remarks today.

Now I will turn the call over to Dr.

Morris Chang, our Chairman and CEO, for his remarks.

Hi, everybody.

This is Morris Chang.

I'd like to make a few general remarks before we start the Q&A.

First, I want to talk about customers' inventory.

Our customers' inventory remained lean in third quarter.

For most of our fabless customers, if we measure the inventory in absolute dollars, then the inventories were about flat at the end of the third quarter, compared to the second quarter.

Only a few customers' inventory dollars increased in third quarter.

If we measure their inventory in days of inventory, DOI, they continued to decline, and in fact appeared to be low, relative to third quarter historical averages.

As far as IBM customers are concerned, most -- majority of their inventories came down in the third quarter also, and DOI the same way came down.

Relative to third quarter history, most of the IBM customers' DOI is kept at medium to low levels.

In summary, our customers' inventory at the end of third quarter looks lean.

Inventory dollars were flat from the previous quarter and DOI has come down.

End market environment for fourth quarter now appears quite positive and we expect world semiconductor market to grow by low single digit percentage sequentially in fourth quarter.

We have revised upwards our full-year estimate for the electronic systems units as follows.

PC will grow 2% year over year, as compared to our earlier forecast of a decline by 4%.

Handsets will decline 5% year over year, as compared to our earlier forecast of 9% decline.

Digital consumer electronics will be barely down, within 1% of last year's level.

Based on semiconductor's strong sequential growth in second quarter and third quarter and a further growth in fourth quarter, we now raise our forecast of semiconductor markets to be a decline by 12% for the full year of 2009.

This is a smaller decline rate, compared to a decline of 17% that we forecast earlier.

We expect foundries to underperform the semiconductor market, but by only a couple of points.

Outlook for the global economy in 2010 continues to improve.

We now expect world GDP to recover and be above last year's level next year.

The SC market will take a bit longer to recover, but will get to last year's level sometime between 2010 and 2011.

Finally, we expect 2010 to be a record year for TSMC.

I know that some concerns have been expressed as to whether there'll be surplus capacity as the result of foundries adding capital spending, and there have been questions as to whether growth will come at the expense of profit.

In the past few months, I have talked often about growth.

I want to clarify what I mean by growth.

It means profit growth, and not just revenue growth.

I have stated that ROE at 20% or above is our long-term profit goal at TSMC.

In the past many years, our ROE normally exceeded 20%.

Going forward, we will not lower our standards.

In the future, in good years, such as next year, I certainly expect our ROE to be above 20%.

As Lora has mentioned in her speech, that we have raised our capital spending to about $2.7b for this year.

This increase in capital spending is in anticipation of the growth that will follow.

We have collected some data and did a statistical analysis.

We found that there is a very good correlation between capital spending and growth for us.

Capital spending tended to lead growth by one to two quarters, so with increased capital spending we certainly expect that our future gross profit -- our future growth will follow.

In the real situation, we have been out of capacity in the advanced nodes, which cover 0.13-microns and below -- 0.13-microns and the more advanced technologies.

We have been out of capacity on those nodes for two solid quarters now.

In spite of our redoubled efforts to increase capacity, which started at around June/July, we are still out of capacity.

In fact, for those advanced nodes, we are out of capacity as far as the eyes can see.

This is, of course, the immediate motivation of our capital spending increase.

Since we based our increase in capital expenditure on our partners' forecasts and cross-checked it with our market research, we do not see an over-capacity situation.

And I want to emphasize that this condition specifically applies to us, and I'm not commenting on other companies' capital spending.

The next question is, if other companies have excess capacity, wouldn't price be affected?

The answer is yes, to some extent.

Again, the fact that we are partners with our customers makes the difference.

In a commodity market such as memory, DRAM, flash, if the industry has over-capacity, then everybody's price gets affected.

But in a custom market, which ours is to a very significant extent, our price will be less affected by a general industry over-capacity.

Over the years, we have tried our best to prevent the commoditization of our market.

We try to do it by three things -- technology leadership, manufacturing excellence and the partnership concept.

In the main, I think we have succeeded very well and I intend to continue to do that.

As far as capital intensity goes, I have often felt that it is a very odd ratio.

The denominator is current sales, and the numerator is capital spending, but the denominator does not depend on the numerator.

Although it's an odd ratio, it does have a long-term significance.

It shows the long-term trend of the business we're in.

Since 2001, the growth of semiconductor has slowed down.

As a result, the foundry industry's growth has also slowed down, although the foundry industry growth is still greater than the semiconductor market growth.

As the long-term growth slows down, capital intensity does decrease.

Back in the '90s, capital intensity was around 40% to 50%.

In the last few years, it's consistently less than that, and it's now around 30%.

On a short-term basis, capital intensity will vary.

For instance, this year our capital intensity will be greater than 30%, but that's going to be rewarded by a higher growth next year than the average for the last few years.

Now I want to talk about the 40/45-nanometer technology progress.

Since we reported on that three months ago, we have fulfilled most of the expectations that we had at that time.

However, new problems have also cropped up, but these new problems are logistical problems.

They have to do with the ramp up of new production equipment.

We have had to not only increase the yields, but also have had to ramp up the wafer starts on 40/45-nanometer in the last six months, and the ramp up speed has been unprecedented.

Lots of new tools were brought in and they added a complexity to production, and the problems that I just mentioned that have cropped up recently had to do with tool/chamber matching.

However, as I said, we consider these to be logistical problems, and we are in the process of resolving them very quickly.

I believe most of these problems will be resolved by the end of this quarter.

In spite of tools-related problems, output on 40-nanometer has increased dramatically.

40/45-nanometer revenue wafer revenue exceeded 4% of total wafer revenue in the third quarter and is on track to reach about 10% of wafer revenue in the fourth quarter.

Because of the production -- the tool/chamber matching problems, the 40/45-nanometer problem did affect the gross margin in the third quarter to the extent of about 1.5 percentage points, and because the problems have not been completely resolved in the fourth quarter, the problem will further affect the fourth quarter gross margin to the extent about 2 percentage points.

But we have taken, of course -- we have taken the fourth quarter impact into account in our guidance.

This concludes my prepared remarks, and we can start the question and answer.

So this concludes our prepared statements.

Operator, please open the floor to questions.

At this time, we will open the floor for questions.

(Operator Instructions).

Your first question comes from the line of Pranab Sarmah from Daiwa Securities.

Please proceed.

Thank you for taking my questions.

First question is on, Dr.

Chang, what is your forecast about the growth in semiconductor industry for 2010?

And you indicated the global semiconductor industry might lag the GDP recovery, it seems like, from your previous comment.

Why it is so this time around?

I heard the first question.

Now, let me answer that, first.

I think our current forecast for the 2010 semiconductor market is that it will be up 10%, up 10% from 2009, and what was the second.

Could you please repeat the second part (multiple speakers).

It's related to the same question.

You said that global GDP will reach 2010 equivalents to 2008 level, but semiconductor industry might take a little bit longer time, sometime in 2010 and '11, somewhere in between.

Why so, like, semiconductor industry lags global GDP, in your opinion?

I think, actually, in my mind, I think there is a couple of reasons.

I think that the first reason is that the world GDP is recovering faster than we anticipated.

I think that half a year ago our expectation was that the world GDP would not recover back to the '08 level until '11.

And now, as I said this evening, we think that the next year the world GDP will recover to the '08 level.

The second reason is that I really think that China, the consumption of semiconductors in China has been very strong and stronger than expected.

And, of course, the world GDP covers China GDP, as well.

But the semiconductor in China is proportionately even stronger than the China GDP.

So those two are the reasons why I think the semiconductor market is recovering faster than we expected nine months, or even six months, ago.

Okay, thank you.

And my second question is the LED side.

You mentioned about the investment of [TWD46m] in the LED business.

Could you elaborate like what type of margin expectations you could expect from this business, because even the industry leader like Cree, they have a 44% gross margin that is the highest now, whereas the other chipmakers' margins are always below [30%?]

Do we expect any gross margin in LED?

Well, it's too early.

I think -- well, right now I think the Board is -- the Board of Directors is just in the process of examining an investment proposal made by the management, an investment proposal of [TWD46m] in a pilot plan of LED.

So it's too early to comment on your question.

Got it.

Okay, thank you.

Your next question comes from the line of Randy Abrams from Credit Suisse.

Please proceed.

Yes, hi, good evening.

I want to see if you can talk about the PC segment.

It's doing a lot better than the other end markets in third quarter, but, then again, in fourth quarter, as well.

And wanted to see if you could talk about the drivers of that acceleration through year end, if it's new products for market share, or if some of that relates to 40-nanometer yields.

Elizabeth will answer that question.

She has the data.

Okay, Randy.

I think your question is with respect to fourth quarter PCs growth driver, right?

Yes.

Okay, well, actually we expect PC shipment to be slightly below its fourth quarter seasonality, at about 9% sequential growth, compared to a seasonal average of 13%.

If you look at a couple of players who have already provided their fourth quarter guidance, I think their sequential increase was below seasonal or can be characterized as being modest.

Okay, so it's actually below normal.

Can you actually go through the other segments, what you expect from fourth quarter and also seasonal?

All right.

We also expect -- right, we also expect handsets, the shipment to grow about 10% sequentially, versus the seasonal of 20%.

Of course, the growth in the developed regions will be moderate in the Christmas season.

Emerging market, emerging regions remain still the major growth driver.

But there's a weaker seasonality after China's October Golden Week.

And then, in India, there's also slowing new subscriber growth.

And, in the digital consumer area, we expect the shipment to grow 55% sequentially and this is slightly above the average of 49%, and of course we expect pretty aggressive price promotion that will stimulate the demand in the holiday season.

We think applications such as game, DVD, set-top box, those will be performing quite well.

Okay.

But I guess wanted to ask was, at the TSMC level, or maybe if you could start with what you're seeing in your end markets for fourth quarter.

And I think within your end markets, PCs are actually doing very well, or at least better, even through fourth quarter, so I'm curious about it from a TSMC perspective, what you're seeing in the end market.

Right.

Randy, as you have read our earnings release, we also said in the fourth quarter we expect computer to perform strongly, and so of course PC graphics, PC chipsets will increase.

Communications we think is going to take a pause in this quarter after two very strong consecutive quarters' growth.

And consumer, because we are on the supply chain high end, the upper side of the supply chain, so the consumer-related applications will decline, according to its seasonal pattern.

Okay, thank you.

And I guess a second question, just on operating expenses, I think at the midpoint of guidance it will be down about 100 basis points sequentially.

Could you talk about how that splits out, R&D and SG&A, that decline.

And off of fourth quarter, how do you expect operating expenses to grow into next year?

Operating expenses in third quarter represented roughly 12% of the revenue.

Of that number, R&D expense, roughly around 6% and SG&A, 4% to 5%.

Going forward, and the whole-year patterns, operating expenses roughly representing 12% of revenue.

We believe that trend will continue through the next year.

Our revenue growth, we will continue to invest more R&D, so that part will representing about 7% to 8% of our revenue.

The SG&A will be about 4% to 5%, so all together will be around 12%.

Okay.

Thank you.

Your next question comes from the line of Michael Chou from Deutsche Bank.

Please proceed.

Hi.

Good evening Dr.

Chang.

I want, if you could give the color on depreciation for next year.

Thank you.

Depreciation for next year.

Lora?

As we have not finalized the CapEx number for the next year, so we don't have depreciation yet.

But, as you know, this year depreciation was essentially flat with last year.

With increasing capital expenditure for '09 and maybe go beyond, '10, we believe depreciation will be slightly higher than 2009.

But we do not have a number yet.

All right.

Thank you.

What was it this year?

It was TWD81b this year, and same number for last year.

Thank you.

Your next question comes from the line of C.J.

Muse from Barclays Capital.

Please proceed.

Yes, good evening.

Thank you for taking my question.

I guess first question, you talked about 4x expectations reaching roughly 10% of revenues in 4Q.

I was hoping you could shed some light on what your expectations are for percentage of revenues in 2010.

And what that would mean in terms of new capacity at that node.

Yes.

We did -- now -- we have not made our detailed plan yet.

But first of all, of course, it is going to be considerably higher than 10% next year for 40/45.

And I think that for the whole year it will climb during the year.

And by the end of next year I expect that it will reach about 20% of our total revenue.

That's helpful.

And then I guess, considering that backdrop, what are your initial thoughts for CapEx for 2010.

I heard on the prepared remarks, more aggressive in 300-millimeter spend.

And I guess if you could shed any more light on what the plans there are, as well as what the linearity looks like.

The linearity of the increase?

Sorry.

The linearity of CapEx spend throughout 2010.

Yes.

Lora, you --

Okay.

As we are seeing strong demand on leading edge technologies across the board, not only on 40-nanometer, also 65-nanometer and some 90-nanometer, 0.13 micron as well, so we are adding capacity now.

And we are going to continue adding capacity.

So 2010 CapEx will be most likely front-end loaded.

It actually depends on the then outlook of 2011 as well.

And as you know, early in the year, this year, we pegged our CapEx at 1.5b.

And now we are talking about 2.7b for this year.

And so now Lora's answer is correct as far as we can tell right now.

But if the outlook of 2011 changes, then we may adjust that.

That's helpful.

If I could sneak in one last question.

Again in your prepared remarks you talked about customer inventory being very lean.

I guess considering that backdrop, and I know there's clearly some uncertainty as to what transpires in Q4 in terms of inventory build, but what does that backdrop give you in terms of visibility to Q1.

Does that suggest that we could see better than typical seasonality, or it's still very early?

I think that the first quarter, first quarter is seasonally down.

And if you look at the past seven, eight years, first quarter has been, on the average, down by about 5% from the fourth quarter.

And my feeling right now, I really don't see anything vastly different from that in the forthcoming first quarter.

That's very helpful.

Thank you.

(Operator Instructions).

Your next question comes from the line of Dan Heyler from Banc of America-Merrill Lynch.

Please proceed.

Thank you.

Good evening Dr.

Chang.

I have a couple of follow ups on the questions today as we ran out of time there.

I wanted to follow up, first, on the R&D question.

I know -- I realize it's difficult to give targets for next year.

But as I look at your asset productivity snapback quarter-on-quarter from 1.3 to 1.6, and also the relatively high OP margin, plus ROE now at 27%, that's the second highest quarter in history.

So I'm wondering why there's somewhat of a reluctance to suggest that next year, or soon, you'll be able to achieve record high ROE.

Is there something that -- some risk factor you think to achieving that that I'm not aware of, or what are your thoughts there?

Well, Dan, I made the comment I think mainly in relation to revenue and profit dollars when I said that the next year would be a record.

Now as far as ROE is concerned, that's a -- that involves a couple of other variables.

And I really haven't taken the time to look at that.

But apparently you have.

And if your calculations indicate that the ROE will be record, well, I will gladly accept it at this point.

Okay, thanks.

And then the other question -- well I guess there's another element there in terms of, of course, the dividend payout as well.

I guess that would be one element there.

And on that front, either for Elizabeth or Lora, could you give some color on policy.

Historically you've talked about dividend payout as a nominal value increasing.

But any thoughts on a payout ratio, to give better clarity as to what your dividend would be?

Yes.

Well, Lora, this relates to dividend.

Right, right.

Dan, you know we had a commitment.

And we are very serious about maintaining our cash dividend level in the past few years.

We have said we will pay no less than TWD3 going forward.

We are still in that position.

Okay, great.

And then finally --

But remember now we actually raised the dividend to this very high level.

As I figure, I guess our dividend yield is now 5%.

And that's a very high level for a company.

We actually -- I actually was rather unhappy about it because we really, when we raised the dividend to that level, we didn't feel that we should invest in capacity more.

And now I think this year is different from the circumstances a few years ago.

And also, we are also reserving some cash -- although we don't really know how much we reserve yet -- for the new businesses, because the new businesses really haven't started to invest yet.

So, but we still feel very good about the TWD3 level.

And we intend to at least maintain it.

Okay.

I just was thinking it was somewhat easier to model as a percentage of quarter revenues, for modeling purposes.

But I fully understand your policy there.

My second question was on the comment about semiconductor industry reaching '08 levels by 2010.

Under -- I guess, at TSMC, there I'm wondering if you could be a little more specific as to -- because obviously second half of '08 was down.

So at what point do you think TSM would be back to, say, first half '08 levels?

Or the semi industry, either way.

Whatever works for you.

Well, when do we return to the first half level of '08?

That was your question?

Yes.

I mean roughly you were running -- TSM was running at about TWD88b at that point.

Well, I mean that will require a quarter by quarter -- just quite simply indicating -- I really thought by simply indicating that next year will be a record year, I really think that I have said enough for the time being, Dan.

I really don't want to -- in fact we haven't analyzed, we haven't planned the next year quarter by quarter yet.

And you really are asking me to give it to you quarter by quarter, if not month by month.

I can't do that.

But I will still stand on my statement that I expect next year to be, on a whole year basis to be a record year.

Right.

Okay, fair enough.

It just seems like that 10% number was a little low on that scenario.

But I'll come back later to you guys specifically on that.

Thanks again.

Thank you.

Well, Dan, if you have more questions you can ask now, otherwise we think we can conclude the conference call today.

Well, operator, since there appear to be no other questions, I think we can conclude today's conference call here.

Before we conclude TSMC's third quarter 2009 results webcast conference call today, please be advised that the replay of the conference call will only be accessible through TSMC's website at www.tsmc.com.

Thank you, all.