TAL Education Group (TAL) 2015 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to the TAL Education Group earnings fiscal-quarter 2015 earnings conference call.

At this time all participants are in a listen-only mode.

There will be a presentation followed by a question and answer session (Operator Instructions).

I must advise this conference is being recorded today, January 22, 2015.

I would now like to hand the conference over to your first speaker for today, Ms. Mei Li.

Please go ahead.

Thank you all for joining us today for TAL Education Group's third fiscal-quarter 2015 earnings conference call.

The third fiscal-quarter earnings release was distributed earlier today and you may find a copy on the Company's IR website or through the newswires.

During this call you will hear from Chief Financial Officer, Mr. Rong Luo.

Following his prepared remarks, Mr. Luo will be available to answer your questions.

Before we continue please note that the discussions today will contain forward-looking statements made under the Safe Harbor provisions of the US Private Securities Litigation Reform Act of 1995.

Forward-looking statements are subject to risks and uncertainties that may cause actual results to differ materially from our current expectations.

Potential risks and uncertainties include, but are not limited to, those outlined in public filings with the SEC.

For more information about these risks and uncertainties please refer to our filings with the SEC.

Also our earnings release in this call includes discussions of certain non-GAAP financial measures.

Please refer to our earnings release, which contains a reconciliation of the non-GAAP measures to the most directly-comparable GAAP measures.

I would like to turn the call over to Mr. Rong Luo.

Thank you, Mei, and thank you all for joining us on our earnings conference call for the third fiscal quarter of 2015.

We are pleased to report on a quarter with revenues and bottom-line results in line with our expectation.

Today I will discuss our operational progress for the quarter, followed by further analysis of the financials and our business outlook.

We maintained strong momentum during the third fiscal quarter.

Net revenue increased 35.1% year over year to $99.4m.

The revenue growth was primarily driven by a strong 35.6% increase in enrollments.

Year-to-date net revenues have increased 37% year over year and we remain well on track to achieve our full fiscal-year projected growth target of approximately 35%.

As I explained last quarter, Q3 was a quarter that we saw the double impact of the late fall term classes against last year and one cost delay to Q4 due to the Asia Pacific Economic Cooperation, or APEC forum in Beijing.

This impact was around RMB30m.

That is 30m in the quarter as we had expected.

If we add back this amount to our actual third-quarter revenue third quarter top-line growth on a normalized basis would have been over 40%.

Let me now provide more detail on each of the three business segments.

Our core small-class offering continued to be the main driver of our growth, up by 37.4% in the quarter.

Enrollment growth for small class in the third quarter was 32.8%.

ASP for small class was up by 3.4% year over year.

Small class contributed 82% of the third-quarter revenue, up from 81% in the same quarter last year.

I am very pleased to know that in terms of revenue contribution of small class, cities other than Beijing and Shanghai for the first time contributed over half, or 51%, of small-class revenue compared to 40% during the same period last year and 48% last quarter.

The ongoing strong performance in cities other than Beijing and Shanghai reflects our success in establishing our differentiated tutoring facilities in a rather favorable competitive environment.

As in previous quarters, a good number of cities achieved triple-digit revenue growth.

The cities include Nanjing, Chengdu, Zhengzhou, Suzhou, Chongqing, Shenyang and Taiyuan.

Combined small-class revenue from Beijing and Shanghai sustained its low double-digit, year-over-year growth, although the retention rate also showed improvement from the last years fall term.

Shanghai recorded high double-digit growth once again this quarter, even as it was negatively impacted by the late fall term.

The enrollments in elementary grades one to three are larger than those in grades four to six, which in turn are larger than junior and high enrollments.

The largest enrollment base is such in the lowest age group this pyramid-like enrollment structure is highly beneficial to maintain Shanghai's growth momentum.

In addition to strong enrollment growth in the fall term, winter enrollments so far are also very good for Shanghai and we expect this trend to continue.

Small-class revenue from Beijing in the third quarter was slightly down year over year.

If there had been no class rescheduling due to the recent APEC forum or the late fall term, small-class revenue from Beijing would have achieved low double-digit growth, driven by ASP increases of over 10% year over year in Beijing.

Due to the ongoing uncertainty in Beijing's exam policy we continue to have declining English enrollments.

The proposed new policy focuses more on conversational English, which fortunately is what we have begun to offer last year in cooperation with the Cambridge University Press under the name Hello English.

And we are pleased to see that half of our Beijing primary English enrollments are by now for Hello English.

And in cities outside Beijing we see that the winter and spring English enrollments are growing by over 50%, that is five-zero percent year over year.

Meanwhile, with our continuous efforts to upgrade the curriculum of Chinese subjects in Beijing we saw enrollments in Chinese subjects increase over 30% in the fourth term year over year.

And these enrollment numbers look good in winter term based on current registered enrollments.

Looking ahead, overall winter enrollments for Beijing point to a modest year-over-year growth with improved utilization and retention.

Let me now turn to the second-biggest segment, our one-on-one business.

As before, we manage the growth of one-on-one as complementary services to our core small-class offerings.

The third quarter is traditionally the weakest quarter of the year as there is no major testing period for students to cram for.

One-on-one revenue grew by 16% year over year and representing 14% of total revenues against 16% in the same quarter last year.

Enrollment growth was 15% year over year.

ASP for one-on-one increased by 1.3% year over year.

The third-biggest segment, online courses, was again our fastest-growing segment with 72% year-over-year revenue growth in the third quarter.

Online contributed 4% of total revenues this quarter, up from 3% in the same year-ago period.

Online enrollments increased 55% year over year.

Online ASP grew by 10.6% year over year.

Our online business was profitable for the fourth quarter in a row.

Online enrollments were 19% of total enrollment this quarter versus 17% in the same year-ago period.

You should keep in mind that the proportion of the online enrollments is relatively large for this quarter because the late fall term and the APEC forum caused a shift of some offline enrollments, mostly in Beijing, to the fourth quarter.

On the absolute basis online enrollments have continued the accelerated growth trajectory we have seen in this fiscal year.

Let me update you now our geography footprint and learning center network.

We see promising enrollment growth in the four new cities we added in the first half of the fiscal year, Jinan, Qingdao, Shijiazhuang and Changsha.

Once again we execute on our proven approach here -- approach to first establish our brand through excellent student outcomes in the first center and, hence, create strong demand based on word-of-mouth before opening new centers.

Our learning center network expanded to a total of 289 centers, a net increase of 2 centers quarter to quarter.

Of our total 289 learning centers, 200 were small-class learning centers, including four learning centers for Mobby and 89 were for one-on-one, whereas, six new small class centers opened and six closed.

The number of small-class centers remained unchanged from last quarter, while we opened a net of two one-on-one centers.

We continue to add center capacity for small class in existing centers with a net of 64 net new classrooms, mostly in Shanghai, Qingdao, Nanjing, Wuhan, Xi'an and Changsha.

Year to date we have added 533 small-class classrooms, an increase of 24% compared with the start of the year, which is in line with the capacity expansion of 24% we achieved for the first nine months for the fiscal 2014.

At this rate we steadily expand our capacity in support of our growth and at the same time optimize center utilization.

For the third quarter we saw overall utilization also improve by 2% as compared with the previous fall term.

In the winter we will continue to add capacity in cities where the utilization is strong, such as Guangzhou.

While in the past year we have spoken a lot about our online strategy, most of our ongoing initiatives are in effect o-to-o, online to offline.

The key logic behind this strategy is for us to drive our core business through integration of our online and offline efforts and get better operational leverage from online to offline.

Going forward we will continue to focus most of our resources on these o-to-o efforts that we will have our offline core business.

To start with the most obvious o-to-o project is Jiazhangbang where we have a large and lively online and mobile community of parents whose children are taking or will take classes in our nearly 300 centers.

Looking ahead we want to acquire more students through Jiazhangbang, which is the most cost effective way for us to grow enrollments.

By the end of November 2014 Jiazhangbang already had over 6m registered users with local websites in 26 cities.

Moreover, we want to further boost our offline business and customer services by adding more online capabilities.

Our ICS 3.0 program is already a clear example of that as well as the Xueersi Peiyou app for small-class registration and blended learning.

In addition we are piloting ways to offer better service such as e-class card and online payment.

Last, but not least, since we expect online learning to be an opportunity for the long term as we are we will also invest in the areas of the online content and online life class platforms.

Finally, before I move onto financial review I would like to briefly update you regarding our investment progress.

We announced in December that we have taken a minority stake in Guokr, a popular mobile and web-based community for science and technology education in China.

Guokr's massive online open course, the MOOC content, will be an important addition to our blended learning approach across multi-media platforms.

Guokr's customer base of high school, college students and young graduates is also highly complementary to our strong K-12 user base.

Our investment in Guokr is therefore an outstanding opportunity to share our vision and align our interest in building new education models for the future and incorporating self-education on the Internet and mobile Internet.

To summarize, the third-quarter result has brought us well within range of achieving our full-year gross target of 35%.

As before, while we drive growth through expansion and technology-based innovation we continue to look for the right balance between present and future growth.

We will allocate our resources both to our o-to-o initiatives as well as continuous center capacity expansion to meet the strong demand for our class-based tutoring.

As for third-quarter financials, I would like to make some brief key points to help explain the numbers as well as give you some directional forward-looking pointers where I can.

Gross margin was 50.7% as compared to 51.2% in the same year-ago period.

Without the postponement of the third-quarter classes to the fourth quarter the gross margins would have been higher on a normalized basis versus the same year-ago period.

Total operating expenses increased by 58.8% year over year to $42.2m due to our stepped-up spending on new business initiatives during the course of this fiscal year.

For the fourth quarter we expect operating expenses to increase from the third quarter within a manageable bandwidth mostly because of the year-end bonuses, adding new capacity and our o-to-o efforts that are estimated to cost up to $15m for the full fiscal-year 2015.

On the ASP side the year-over-year blended ASP was almost the same for the third fiscal quarter from the same period of the previous year, which continued to reflect a very positive shift in our business in favor of small class and online as we manage the growth of our one-on-one business.

Earlier in my prepared remarks I already gave you separate ASPs per business segment because blended ASP no longer properly reflects the underlying trends due to the shift in business.

Let me briefly recap these ASP numbers.

Small-class ASP grew 3.4% in the quarter, reflecting price increases in 30 cities we mentioned previously as compared to the same period of the prior year.

Next year we will continue to increase small-class prices in some selected cities.

ASP for one-on-one increased by 1.3% year over year in the third-quarter, fiscal-year 2015.

The ASP for online courses in the third quarter increased by 10.6% year over year.

Income tax expense was $400,000 in the third quarter of fiscal-year 2015 as compared to $1.9m in the third quarter of fiscal-year 2014.

The decrease was mainly because one of our entities, TAL Beijing, has qualified as a high and new technology enterprise strongly supported by the State in the third quarter of fiscal-year 2015.

Therefore, it's entitled to have a preferential tax rate of 15% from calendar-year 2014 through 2016.

We tune up the impact of around $1.1m of the first half in the year into the third quarter.

GAAP and non-GAAP income from operations declined by 30.9% and 6% respectively year over year.

Basic and diluted net income per ADS were $0.14 and $0.13 respectively for the third quarter.

Non-GAAP basic and diluted net income per ADS, which excludes share-based compensation expenses, were $0.20 and $0.19 respectively.

Let me finally turn to our guidance.

In the fourth quarter we again deal with the usual every other year seasonality associated with the timing of Chinese New Year.

The late start of spring term, which typically begins after Chinese New Year, will cause one less weekend of classes to be recorded in February as compared to last year.

Taking into consideration this seasonality impact and the revenue shift from Q3 to Q4 we mentioned earlier, for the fourth quarter of fiscal-year 2015 we expect total net revenue to be between $116.6m and $119.2m, representing an increase of 34% to 37% on a year-over-year basis.

In addition, please note that this guidance includes the quarter-to-date exchange rate impact, which is negative 2% as of today and assumes no more material change in the exchange rate.

For the fiscal year ending February 28, 2015 we expect total net revenue to be between $427.4m and $430m, representing an increase of 36% to 37% year over year.

These estimates reflect the Company's current expectation, which is subject to change.

That concludes our prepared remarks and, operator, we are now ready to take questions.

(Operator Instructions).

Jialong Shi, Credit Suisse.

Hi, good evening, Rong and Mei.

Thanks for taking my call.

First of all, congratulations on a very solid quarter.

I notice both your marketing and G&A expense increased quite significantly in Q3.

Could you provide some colors on the drivers for these two items, and also just wonder if you could provide some guidance for your OP margin in Q4 and next fiscal year?

Thank you.

Thank you, Jialong.

I think as what we discussed in our earnings call last time we have [highlighted it] because we are adding around 300 technical staff, both the IT engineers and product managers we have [integrated] into our organization starting from Q2.

So especially in Q2 some people they are coming the first month, some people come into the second month, but in Q3 they have a full quarter.

So if in a quarter-to-quarter perspective actually our OpEx increase around $2m to $3m that is -- most of that is because of their full quarter in Q3.

So looking forward to Q4 and probably next year I think, same as before, we don't provide a guidance on gross margin or operation margin in the coming quarters, but I can give you some of the directional points from our current visibility.

For Q4, as you'll notice in our guidance, the net impact of revenue shift from Q3 is almost offset by the every other year scenario associated with Chinese New Year.

Also we have already seen there will be some impact from the appreciating dollars.

Quarter to date it's around 2%.

So even so we still expect our Q4 year-over-year revenue in top-line steady to still grow 34% to 37% and small-class revenue growth in renminbi terms will still be very good, over 40%.

As you have mentioned discussing my Q3 results, we also -- in Q4 we still expect our operating expenses to grow from the third quarter within a manageable bandwidth, mostly because of three things.

Number one is the year-end bonuses.

Typically, Q4 is the big quarter where we give our management and employees their main bonuses.

The level of that -- the level of the bonus I think you could probably benchmark what happened in last year Q4 versus last year Q3.

The second thing we are also doing is we are adding new capacity starting from Q4.

And the last one is our o-to-o efforts, including Jiazhangbang, including Peiyou app, including the live online and ICS 3.0.

All of these o-to-o efforts are estimated to cost up to $15m for the full fiscal-year 2015.

So directionally I think our non-GAAP operation margin in Q4 will be below same quarter last year because of all of these o-to-o investments just happening this year.

But compared to Q3 we will still be a little better than Q3.

Looking forward to fiscal-year 2016 I'm sure most of you will be very interested.

What I can say is we will continue to look for the right balance between the top-line growth momentum and the investment for future growth as we have managed in the past.

And therefore I think in the coming years we are still confident we can maintain the sustainable and healthy top-line revenue growth while the non-GAAP operating margin will be moderately down, or in the best case remain flat for the fiscal-year 2015.

Since we are still in the budgeting season so we haven't finalized our budgeting yet, so that is -- I hope I can give you more FY '16 color during next quarter's earning call.

So that is what -- that's based on what I can see today.

I hope it can help to answer your questions.

Thank you for the detailed answer.

Thank you.

I'll get back to the queue.

Thank you, Jialong.

Fei Fang, Goldman Sachs.

Hi, good evening, Rong and Mei.

Thanks for taking my question.

Can you update us on your expansion plan over the course of 2014?

How many centers would you plan to add, what are the geographical focus and would you plan to expand into more cities?

Thanks.

Thank you, Fei.

Continuing to adding more capacity is always our priorities.

Year to date in the first three quarters we have added 15 learning centers.

In classroom perspective we have added 533 classrooms by the end of November.

In Q4 this year we will be a little different as we start to add more capacities one quarter earlier than before.

So in quarter four we have seen some of the cities they -- with the very high utilization rate, so we try to add more over there.

In all over quarter in Q4 I think we probably will add the other five to seven learning centers in Guangzhou, Chengdu, Wuhan, Zhengzhou and other cities.

And the net increase of the classroom will be around 100 to 150.

And on the other side, which is a little different than before, is we don't favor the small learning centers now, so we try to open some big learning centers in the future, which is the most efficient way to optimize our investments there.

For example, in Guangzhou the new centers will be over 2,500 square meters.

And the centers in Wuhan maybe were over 3,000 square meters.

The centers in the others, in Chengdu and Zhengzhou and other cities were over 1,000 square meters.

And looking forward to next year, fiscal 2016, I think in general we'll still maintain our pace to add another around 20% to 25% in capacity perspective.

We still have some priorities when we try to add capacity.

Number one is we have a close look at the utilization rates in different cities.

So we will monitor that metric and try to add as much as possible.

The second thing is our cities we have just added in the year 2012 and the year 2011 they have reached a very important milestone and we will carefully be looking at their needs and try to act more in the future.

At the same time for some of the cities, like Beijing, we will also try to do some optimizations.

We will consolidate some of the very small learning centers to one big learning center, which has been proved is a much more efficient way to our students and teachers.

And on the other side I think I also would like to give you more information about capacity perspective is our plan to adding new cities next year.

Again we will maintain our pace to adding maybe two to four new cities next year.

By the end of today we have 19 cities in our hands, but we still have a lot of big cities with great potential in our wish list, for example, like Dalian, Ningbo, Fuzhou, Xiamen and etc.

So all in all I think we will continue to execute our plan to add more capacities in the relatively stable pace.

And we will let you know and update you more details when we come into the end of each quarter.

Thank you, Rong.

That's very helpful.

Thank you, Fei.

Philip Wan, Morgan Stanley.

Hi, good evening, Rong and Mei.

This is George calling on behalf of Philip.

Thank you very much for taking my questions.

So my question is regarded -- regarding the pricing of your courses.

I understand that it's because of the mix shift so the overall ASP is down, but -- or this is flat.

But even if we look at just the small-class ASP overall it's still only 3% this quarter versus maybe 6% last quarter.

I know in some cities, for example, Beijing, the ASP is still up 10%, but do you see more competition that is driving -- or that is preventing you from increasing the price across the country going forward?

We noticed that New Oriental actually introduced some loyalty program recently that effectively lowered their pricing by about 5% -- 2% to 5%.

So what do you think of that?

Do you think that the competition is actually heating up so you also probably need to give some discount or to slow your pace of increasing the price?

Thanks.

Thank you, George.

The simple answer is no.

I think our core small-class ASP we still maintain a very healthy increase, which is 8% -- close to 9% in Q1, 6% in Q2 and 3.4% in Q3.

This year we took increase in Beijing, Shanghai, Shenzhen, Wuhan, Zhengzhou, Suzhou, Chongqing and Taiyuan since the same period last year and we plan to do so in selective cities next year as well.

So we will continue to increase our price from time to time.

And one thing I need to remind you is our geography distribution moves to cities other than Beijing and Shanghai.

Just as I mentioned in my script, the cities outside Beijing and Shanghai now contribute more than half of the revenue in my small class now.

And in general the price in the outer cities is a little lower than the price in Beijing and Shanghai.

So that will be part of the reason why you'll see the slightly lower ASP growth.

And I think looking forward we will still maintain our pricing power over here.

We can still expect to see our small-class ASP growth will be around 3% to 5%.

And at the same time for our one-on-one business online ASP will remain stable in the coming quarters.

Okay, very helpful.

Can you -- quickly can you remind us in the outer cities what's your average premium -- price premium for your small classes versus the local player, the competitor?

Thanks.

Because it's different city by city, but what I can say is we can use some big numbers to do the benchmark.

In general I think we can say around 30% to 50% price premium over our local competitors.

Okay, got it.

That's very helpful, thank you.

Thank you, George.

Leon Chik, JPMorgan.

Yes, hi.

First, congratulations, and also I was wondering if there's any color you can give on maybe other courses other than math and science, any expansion plans into those courses, maybe more humanities or the languages etc.?

And in -- if that's the case, or these are just not things that you're interested in?

Thanks.

Hi, Leon, can I clarify your questions?

Do you want to understand our progress in maybe English and Chinese in more cities?

Yes, English and Chinese in more cities and also maybe just to add there are probably quite a lot of courses that you're not offering right now.

Yes.

Just expand your scope of your core -- your math and science is what you're good at, but is there any -- I'm trying to see because it looks like the growth in cities and maybe even the tuition may be not as fast as --

I understand.

-- some people want, and if there is another avenue of growth, even post-secondary, for example.

Okay, okay, okay, I fully understand.

So specifically in English and Chinese I think we are making progress in English.

As I mentioned just now, the winter term and spring term the enrollment growth in English in outer cities other than Beijing is over 50%.

In winter it's over 60%.

In spring it's over 50%, but we still have three weeks to go too, so that number will go up definitely.

And we also expand our cities with the English class maybe from five to seven.

So I think for English we find our conversational product Hello English is a good product and we try to expand more in our existing cities we have math and science there.

So that definitely will be a big plus.

For the subject of Chinese we also were pleased to see in Beijing -- Beijing today is the only city we have Chinese.

So we are also pleased to see in Beijing the Chinese enrollments also grew more than 30% while the overall Beijing enrollment is almost flat, so we also feel good about that.

In next year we're also thinking to expand the Chinese to more cities, but again we will be relatively conservative.

We want to make our product right, make our business model right before we go into more cities.

Okay.

Thank you, Leon.

Vivian Hao, Deutsche Bank.

Hi, Rong and Mei.

Thank you for taking my question.

I've got two questions here.

First of all, can you tell us what is the current average utilization rate of your classrooms and also maybe the benchmark of what kind of utilization once you hit that benchmark that you will expand your capacity?

Second question is regarding your o-to-o initiatives.

Could you elaborate a bit more on the progress, especially your plan of investment for online- and mobile-related areas for next year?

Thank you.

Okay.

Thank you, Vivian.

So for the utilization rates in different cities, they are quite different.

I think, for example, in Beijing they're relatively -- it's a little lower because in Beijing we have more subjects, like Chinese and English in Beijing.

And for some of the cities, like Guangzhou or in the other places, they are much higher, more than 90%.

But in general frankly speaking when we're looking at to -- looking at the utilization rates I think 60% is a very important benchmark for us, because 60% means 80% of the classroom and 80% of the fees are also paid at the same time.

So when our learning centers hit 60% we will evaluate the need to add more capacities in that learning center or in that city.

For the o-to-o efforts, as we have mentioned in my prepared remarks, I can give you more details about that -- the four projects there.

For Jiazhangbang, which is a very important tool for us to acquire new customers, the registered users of Jiazhangbang both the PC and Web have surpassed 6m registered users.

And one of the good things we are seeing today is the UV who has logged on the Jiazhangbang app is higher than the people who log on the Jiazhangbang Web.

And we are also seeing the duration of stay in the Jiazhangbang app is also longer than the Jiazhangbang Web.

So now we have the local websites in 26 cities which enable us to try a lot of new business models and try to promote our branding and our core offerings more efficiently.

The second thing is about our Peiyou app.

Peiyou app today -- in October, that's the first month we used Peiyou app into market.

We pilot in 10 cities.

We used the Peiyou app to do the registration and to do the online payment.

The feedback from the Peiyou app is quite positive.

In some of the selected cities we even tried a model to close all of our service centers and the feedback's also very good.

And by the same time we also see some of the small technical bugs coming from there, because our peak volume of the registration is quite high.

So we still need one or two terms to improve the functionality of Peiyou apps, but I personally feel very good about the progress we are making today.

I will share more information when we come into summer term.

And the third thing is we've talked about the ICS 3.0.

The ICS 3.0 we have deployed 3,000 in Q1 and the feedback from the teachers and students is also very positive.

That's why we will continue to deploy more starting from Q4.

And next year we also have the plan to deploy this system into more cities and into more subjects and into more grades.

And the last one is about the live online Haibian.

Live online Haibian is also making good progress there.

Now we have over 1,000 students, a couple thousand enrollments now.

When we just started this platform we only had maybe 100 students there.

So for Haibian we will still treat it as pilot stage because we want to make the product right and we -- now we have extending the Haibian pilot from three cities to seven cities now.

And that seven including Beijing, Guangzhou, Shenzhen, Xian, Zhengzhou, Jinan and Changsha, so we will continue the investment in Haibian live course project to improve its functionality and try to expand more grades before we officially promote in the country wide.

So all in all I think we feel good about the progress we are making today.

I think we feel confident they will give us long-term return with the time moving on.

This is very helpful, I just have one quick follow up.

Does that mean that the investment magnitude will be similar to this fiscal year -- sorry, to FY '15 for FY '16?

That's different project by project.

I think the investment of the Peiyou app and Jiazhangbang will be moderately down or even flat coming to next year, because it's almost there.

The investment -- the content development for ICS 3.0 is almost there, but we need to consider because we want to deploy more tablets to more cities, so we need to consider the depreciation costs from there.

Haibian is still in the pilot stage.

Even their growth is quite healthy and quite exciting, but we still treat it as a pilot stage.

We will continue to invest more over there.

So in general I think when I finish my fiscal-2016 budget I can share more about that.

Thank you.

This was very helpful.

Thank you, Vivian.

Clara Fan, Jefferies.

Hi, hello, thank you for taking my question.

I've got a follow-up question in regards to your English class.

So I understand that it's only offered in a few cities.

So in terms of revenue contribution how big is it now for our English classes?

And also I would like to ask what's our dividend.

I understand our dividend policy is that we don't regularly give out dividend.

I'm just wanting to have a feel on what's going to happen this year.

Thank you.

Thank you, Clara.

For English, actually, all over the country is around 10% to 12% of my overall revenue.

In Beijing it's higher; it's close to 15% of my overall revenue.

I think this percentage will go up in the future.

And about the -- especially about the dividend, same as before, we will continue to make strategic investments to support our vision of becoming a leading technology-focused education service provider.

So we are making some small investments and we will continue to make more investments with the synergy of our core business, which is a little different from this year.

Year to date we have invested some of the deals and all of them are less than 20% minority stake.

In the future we will focus on our K-12 business.

We will focus on our core business.

And we will look at the deals with high synergy with our core business.

And we expect the number of the deals will be down next year and maybe the average size of the deals may go up.

So we still want to leverage our cash reserve to make some strategic investments in a controllable way.

So we are unlikely to do a dividend in the near term.

We will balance the long term growth into our considerations.

Thank you, just one follow-up question.

So how about math?

How much does it account for our revenue then?

Math is over 60%.

Okay, thank you very much.

Thank you, Clara.

Ella Ji, Oppenheimer & Co.

Thank you for taking my question.

First question is relating also to your operating expenses.

Can you give us an update for the key cost components, such as the teachers' compensation, and also our rent expenses?

How much year-over-year increase are you seeing there?

For OpEx the teacher compensation and the rental the percentage of net revenue is still stable.

The year before last year it's around 21%.

Last year it's around 20%.

This year it's around 18% for the teacher compensation.

Rental is also quite stable.

Okay, but if you are saying this year is 18% is it indicating a declining as a percentage of revenue, a slowly decline?

Yes, for the teacher compensation it's slowly declining around maybe one point per year and the rental is around 12% to 15%.

It's different quarter by quarter, because, for example, in Q3 and Q4, especially in Q3, we start to add more capacity there.

So in that quarter maybe it will be around one or two points higher.

But if normalizing a one-year perspective is also quite stable.

Got it.

Thank you.

And then second relating to your investments for the online initiatives, I wonder if you can also help break it down for your budget for maybe next year.

Are we going to see -- for example, which line in the operating expenses are we likely going to see more increases?

Is it going to be more in the sales and marketing, or in the G&A, or in the cost of the revenues?

More than 60% will show in the G&A.

The rest will spread in the S&M and costs.

Got it.

Thank you.

Lastly, could you also talk about -- so far you have made a lot of investments in -- as a minority shareholder in other websites.

Could you talk about what type of synergies have you seen so far?

And is it likely that you will maybe increase your investment to be a majority shareholder rather than a minority shareholder?

Yes, that depends the development of our -- the deals we invest.

I think we need to consider that in a bigger environment.

Now we're looking into the online education, where we all know the online education today is still pre-revenue and we don't see very good or very mature models in this market yet.

But at the same time we're also seeing for the online education new projects there we are seeing they can quickly get a lot of traffic over there.

So some -- they are also very helpful and beneficial for us to expand the consumer courage.

Before, most of our students coming from primary school and K-12, but with the online education we even can touch more consumers.

For example, when we invest in Guokr they provide us an entrance to touch the high school students, college students and the graduate students.

So the reason why we take the less than 20% minority stake is, frankly speaking, we are taking a relatively conservative approach because we feel online education in future definitely will have a bright future.

But now we are still -- we don't have enough information.

We are not smart enough to judge which deal and which project will succeed in the future.

We take a relatively conservative approach to have less than 20% in stake, but we keep a very close watch of what they are doing today.

And we're also sending our senior people to become their Board directors.

When we see some of the good things, good signals coming out we will increase our stake to a higher level, maybe become majority.

But at the same time if after several time we see the deals actually don't have any synergy with us, or maybe they are good but they don't have any good correlation with our core business, it's purely a financial investment with us then we will decide to maybe adjust so our stake or we just quit.

So I think we still need several quarters to evaluate and keep close watch of what's happening in this market yet.

Got it.

That's very helpful.

Thank you, Rong.

Thank you, Ella.

Tian Hou, T.H. Capital.

Hi, Rong and Mei.

I have two questions.

One is your -- related to your investment.

Regarding the o-to-o investment, I realize there are some investments can be one time, such as buying equipment or software outsource, and some other investments could be the human capital investment, which could last into next years.

So I wonder this year how much are those o-to-o additional investments?

How much of that will go into -- will continue into next year or two and how much of that will not happen in the next year?

That's the number-one question.

Number two is on your income statement.

And there is one question maybe because I'm not accounting background, so I don't really quite get it.

Last quarter your share account fully diluted is 90.6m shares.

And this quarter there was a 9% decline, become 82.4% -- 82.4m, and where you have a 2m share account deduction.

So I wonder what's the reason?

And what's the rule of accounting?

And going forward how should we model this part of the item?

That's all two questions.

Okay.

Thank you, Tian.

Especially to your first question, I think today our o-to-o investments in large part is headcount, it's the people.

By the end of November this year we're adding maybe incremental around 300 staff, both IT engineering and product managers there.

And looking forward to next year we don't see we will increase that much.

But the people we hire we also don't have the plan to quickly fire all of them, so they will still be there.

And in cost perspective they -- we will still need to assume maybe a fair salary increase.

For the small part it's the investment we're spending to do the project -- outsourcing projects on some of the devices and facilities.

So all of them will be amortized in three years.

So we don't have similar type of this kind of facility investment next year.

Specifically to your second question about our EPS, I think that is more like an accounting treatment.

According to the financial accounting standards the calculation of the diluted EPS should not assume conversion or asset size or contingent insurance of securities that will have an anti-dilutive impact on EPS.

In determining whether potential common shares are dilutive or anti-dilutive each issue or series of issues of potential common shares should be considered separately, rather than in aggregate.

Which means our accounting treatment of the convertible bond is based on the if-converted method.

When we calculate the diluted EPS we should consider two scenarios.

Number one is we're assuming CB will have an anti-dilutive effect, so it's the impact to net revenue and the share counts will not be included in the calculation.

Which means the first scenario -- the formula will be EPS equals net income divided by [diluted ADS, including the impact of stock incentive grants from share-based compensation] (corrected by company after the call).

The second scenario is we're assuming CB will have a dilutive impact on EPS, all bonds are converted, the interest expense incurred from the CB around $2m Q3 should be added back to net income.

And the share accounts from the convertible bonds will be included into the denominator.

Which means the formula will be EPS equals net income plus interest expense, divided by [diluted ADS, including the impact of stock incentive grants from share-based compensation, and convertible bonds] (corrected by company after the call).

So we -- from a cost perspective we need to compare these two and we need to choose the lowest EPS to report on.

So that is the accounting treatment.

And going forward how should we -- it's kind of -- for me, it's complicated, so I want --

Yes.

-- forward, where a simpler version or guidance how should we model going forward that?

Pretty simple, if you are modeling a full-year number maybe the $90.6m is better.

If you are modeling a one-quarter number and that quarter number is -- the net income of that quarter number may be not that much, then you may use the lower numbers.

So that's purely accounting treatment we have to follow US GAAP.

Okay, that's all my questions.

Thank you.

Thank you, Tian.

Okay, thank you, all of you, to join our conference call.

We know it's quite late in China now.

And I'm looking forward we have more conversations after call.

And, again, we are looking forward both Q3 and in the coming year I think from Company perspective we will still continue to deliver the strong top-line revenue growth.

And we will also try our best to maintain the relatively stable gross margin and operation margin in the future.

So thank you so much for coming and looking forward have more conversations in the future.

Thank you.

Thank you, sir.

Ladies and gentlemen, that does conclude our conference for today.

Thank you for participating.

You may all disconnect.