TAL Education Group (TAL) 2018 Q1 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to TAL Education Group First Fiscal Quarter 2018 Earnings Conference Call.

(Operator Instructions)

I must advise you that this conference is being recorded today, Thursday, 27th of July, 2017.

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

Thank you.

Please go ahead.

Thank you all for joining us today for TAL Education Group's First Fiscal Quarter 2018 Earnings Conference Call.

The first fiscal quarter earnings release was distributed earlier today and you may find a copy on the company 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 U.S. 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 the 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 now like to turn the call over to Mr. Rong Luo.

Thank you, Mei.

Good evening and good morning to you all, and thank you for joining us on this earnings call.

I'm pleased to report a strong first fiscal quarter 2018 results.

Our top-line growth continued to be driven by high demand in all cities and supported by the further capacity expansions.

Top-line growth in the quarter was 65% year-on-year in dollar terms, to USD 322 million.

In RMB terms, net revenue grew by 75% year-over-year.

Enrollment increased 62.2% year-on-year.

Income from operations in the first quarter increased by 63.9% and 52.1% on GAAP and non-GAAP basis.

This fiscal year we expect a similar margin trend as we had last fiscal year.

The margins will vary on a quarterly basis as a result of our expansion early in the year, and the time it takes for newly opened learning centers to ramp up.

In the quarter, we entered 5 new cities and added 60 new learnings centers.

After renovations and other preparation work the cumulative classrooms will gradually be used.

As we have seen last year, we expect to benefit from the capacity expansion and see more leverage in the second half of the fiscal year.

Now, Mei will give an update on our operational progress in the first quarter.

After that, I will discuss our continuous investment in technology, which is guided by our long-term strategy vision of advancing education through science and technology.

Revenue growth was supported by the growth of cross-business lines in all cities, including our newly acquired business.

Small class, which consists of Xueersi Peiyou small class, Firstleap, Mobby and some educational programs and services accounted for 81% of total net revenue, compared to 83% in first quarter last year.

Peiyou small class, which remains our core business, represented 71% of total revenue, lower than 73% in the same year-ago period.

This reduction was mainly due to the consolidation of the newly acquired business.

Net revenue for small class was up by 61% in dollar terms and 71% in renminbi terms, while enrollment increased by 69%.

Xueersi Peiyou small class revenue generated from cities other than the top 5, Beijing, Shanghai, Guangzhou, Shenzhen, Nanjing, grew by approximately 88% in renminbi terms.

The renminbi revenue growth in Beijing was around 40% year-over-year in the first quarter, driven by incremental enrollments from the plenty of new classrooms added in previous quarters.

Shanghai gradually recovered, as we had previously expected and achieved year-over-year revenue growth higher than Beijing in the quarter.

Cities other than the top 5 accounted for 39% of the Peiyou small class business, and increased around 35% in the same quarter last year.

We have achieved in renminbi terms triple-digit year-over-year revenue growth in 11 of our top 25 cities that we entered by the first quarter fiscal year 2017, including Shenzhen, Jinan, Shijiazhuang, Qingdao, Changsha, Luoyang, Nanchang, Ningbo, Hefei, Wuxi and Fuzhou.

Let me give you an update of the summer promotions.

In the past two summers we held promotion activities in Beijing.

These promotions effectively turned our Beijing school back into a strong enrollment-driven model and it helped us gain share in the market consolidation.

As a result, this summer we hosted promotions in Beijing and a limited number of other cities.

We also offered more packaged sales of the top courses this summer term in order to gain more market share.

Most of these marketing activities are targeted at first grade of junior high, and other grades to a lesser extent.

We expect these offers will once again help further drive enrollment growth in the coming quarters.

Zhikang one-on-one business experienced a strong seasonality-driven quarter and also benefited from the strong growth of our small class business.

Zhikang one-on-one achieved year-over-year revenue growth of 55% in renminbi terms, based on strong demand and to a lesser extent by normal price increases.

On a net basis, we added 1 one-on-one learning center in each of Shanghai, Shenzhen, Zhengzhou, and closed one center in Guangzhou.

Zhikang one-on-one accounted for 13.2%( corrected by company after the call) of total revenue, compared to 13.1% in the same year-ago period.

Turning to our capacity expansion, in the first quarter we added a net 60 new learnings centers.

During the quarter, we added a net 55 Peiyou small class learning centers, or 1,532 classrooms.

We ramped up learning center expansion earlier in this year.

Because early registration for classes showed that enrollment would be strong, we have opened 3 Firstleap small class learning centers, one each in Beijing, Chongqing, and Shenyang.

This quarter we added most small class classrooms in Shanghai, Xi'an, Guangzhou, Beijing, Hangzhou, Zhengzhou, Chengdu, Qingdao, Shenzhen, Shenyang, Tianjin, and Wuhan.

As planned, we also entered 5 new cities, Dongguan, Changzhou, Xuzhou, Nantong and Foshan.

By the end of May 2017, we have 567 learning centers in 35 cities across China, of which 394 were Peiyou small class, 8 were Mobby small class, 56 were Firstleap small class, and 109 were Zhikang one-on-one.

Looking to Q2, we are planning to add 20 to 25 Peiyou small class learning centers.

We accumulated a classroom capacity early in the year ahead of the peak summer season, and also hired and trained new teachers in advance.

As a leading education brand, we continued to recruit top-level graduates in all of our cities.

The new teachers and classrooms will be gradually utilized and will contribute to our top line.

This seasonality will cause our margins to be under some pressure in the first half of the fiscal year, as Rong already mentioned.

We expect margins to recover during the second half, as we managed last year.

Moving now to our online business, in the first quarter revenue from xueersi.com grew by over 100% year-over-year in renminbi terms.

Online is contributing 4.9% of total revenue this quarter compared to 4.3% in the year-ago period.

We are pleased with the ongoing progress of live broadcasting on xueersi.com, and excited about the long-term opportunities in this field.

We strongly believe that by doing well in innovative technology, particularly artificial intelligence technology, AI, we can drive the online live teaching model to a much bigger potential.

In a few minutes, Rong will further elaborate on this.

Finally, on the revenue side per business line, other revenue is mostly from the online advertising business and represents 1.1% of total revenue in the first quarter.

We would like to remind everyone that as we have announced in today's earnings release on August 16, TAL stock will be split and 1 TAL common share will equal 3 TAL ADS, changing from 1 common share representing half in ADS.

For TAL ADS holders as of August 8, 2017, this ratio change will have the same effect as a 6-for-1 ADS split.

Let me now go through some key financial points for the first quarter of fiscal year 2018.

In the first fiscal quarter, small class ASP in renminbi terms increased by 1.4% year-over-year, mainly driven by the regular price increases.

Zhikang one-on-one ASP in renminbi terms increased by 7.0% because of price increases.

Online course ASP went up by 42.4% in renminbi terms in the first quarter, mainly because of the increase of sales of the online live class, which generates higher ASP than pre-recorded classes.

Both GAAP and non-GAAP cost of revenues increased by 68.8% to USD 169.6 million from USD 100.5 million in the same year-ago quarter.

The increase in cost of revenues was mainly due to an increase in rental costs and teacher compensation.

In the first quarter gross profits was USD 152.3 million as compared to USD 94.6 million for the same year-ago period.

Gross margin for the first quarter was 47.3% as compared to 48.5% for the same period of last year.

Operating income increased by 63.9% to USD 28.8 million.

Non-GAAP operating income increased by 52.1% year-over-year to USD 39.5 million.

The margin seasonality will be similar to last year.

Basic and diluted net income per ADS were USD 0.35 and USD 0.32, respectively for the quarter.

Non-GAAP basic and diluted income per ADS which exclude share-based compensation expenses were USD 0.48 and USD 0.43, respectively.

From the balance sheet, as of May 31, 2017, we had USD 719.5 million of cash and cash equivalents and USD 398.2 million of short-term investments, compared to USD 470.2 million of cash and cash equivalents and USD 229.5 million of short-term investment as of February 28, 2017.

We paid a cash dividend of USD 41.2 million in May.

Capital expenditures for first quarter were USD 29.6 million, representing an increase of USD 16.9 million from USD 12.7 million for the same year-ago period.

The increase was mainly due to leasehold improvements and the purchase of servers, computers, software systems, and other hardware for the company's teaching facilities and the mobile network research and development.

As of May 31, 2017, the company's deferred revenue balance was USD 959.4 million, compared to USD 558.7 million as of May 31, 2016, representing an increase of 71.7%.

Deferred revenue primarily consisted of the tuition collected in advance from the summer and fall semesters of Xueersi Peiyou small classes, as well as deferred revenue related to acquired businesses.

Now I will hand the call back to Mr. Luo to highlight our recent progress in exploring science and technology, and the business outlook of the next quarter.

Thank you.

I would like to share with you our new findings and strategy directions of the future investment.

From year 2019 on, TAL has been exploring the greatest value of the internet in order to better match the supply and demand of quality education resources.

We have recently concluded the allocation class AI could change the future.

For example, we are exploring some new technologies in Xueersi online school, such as the use of face recognition, speech recognition, and touch interaction.

With more interpersonal connectivity between the student and teachers, we can help the students to adopt better study behavior with more diverse online learning closed loops, and a clear learning plan.

The upgraded system brings the online study experience closer to the offline experience and offers unique advantages in many areas.

Early results show it significantly improves the student focus, interactions, as well as productivity and it lets teachers teach based on different aptitudes.

Our exploration of education plus AI is still in early stage.

Considering the complexity of the technology and the characteristics of education, it requires more time to bring this into full play.

We will continue to invest in science and technology, finding new opportunities.

But first and foremost, we will maintain the current healthy growth in both the top line and the bottom lines for the foreseeable future.

Let me now move to the outlook for the next quarter.

Based on our current estimates, total net revenues for the second quarter of fiscal year 2018 was expected to be between USD 428.4 million and USD 433.8 million, representing an increase of 58% to 60% on a year-over-year basis.

If not including the impact from depreciation of the renminbi against the US dollar, the projected revenue growth rate is expected to be in the range of 61% to 63% for the second quarter of fiscal year 2018.

These estimates reflect our current expectations, which is subject to change.

That concludes my prepared remarks.

Operator, we are now ready to take questions.

(Operator Instructions) Your first question comes from the line of Alvin Jiang from Deutche Bank.

I have two questions.

The first one is about margin trends.

There's a rapid capacity expansion in the last two quarters.

Do you think the quarterly and the full-year operating margin will see pressure because of this capacity expansion?

So the first question actually, yes, we have added in the past two quarters.

I think last year Q4(corrected by company after the call) we added around 1,145 classrooms, and this year Q1 we added around 1,532 classrooms.

Compared to the same year-ago period, which is only 305 classrooms and 768 classrooms actually we are adding more classrooms to our current networks.

So for the new classrooms, actually, it generally takes us 2 quarter to ramp up, because they need some time to do renovations and preparations.

And at the same time, we can't wait for classrooms to open to hire teachers.

So we need to prepare teachers around 2 quarters, we hire 2 quarters before that.

So which means this kind of costs we have spent in both the rental and the teachers, will not bring us so much revenue in the coming 1 or 2 quarters, which will come into effect and see much more improvement starting from Q3 in fall term.

So you probably can see my Q1, my gross margin have been declining by 1.2%, which can explain why my non-GAAP operating margin is only declined by 1 point.

Same as what we told to the street, I think, starting from 2 to 3 months ago, this year the margin trend will be quite similar to last year, which means the first 2 quarters, the margin will be a little bit down compared to last year, and while it will recover in Q3 and much better in Q4.

In the whole year, we still maintain margin and non-GAAP operating margin will be flattish.

So we still feel confident about that.

My second question is about the revenue growth.

How big is the revenue contribution from Shunshun and Firstleap in Q1 to Q2, and what's the revenue growth outlook for the full year?

Shunshun and Firstleap, our two important newly acquired businesses last year, I think we start to consolidate Firstleap from February, and we start to consolidate Shunshun only from August last year.

So coming to Q1 and Q2 actually, they're add-in together is around 5% to 6% of my total revenue.

They also growing quite well and align with our average top-line revenue growth from Peiyou perspective.

And for the full year, if you we look into the full year, in Q1 we grew in renminbi terms it's around 75%.

In Q2 our guidance is 61% to 63%.

So we are on the right track to deliver a very healthy top-line revenue growth.

Previously what we said is our top-line revenue growth in the coming three years will be in the range between 30% to 50%, and definitely this year we're much closer to the high end.

Zoe Zhao from Credit Suisse.

I have two questions as well.

The first question is regarding the utilization rate after we added this many capacity.

Could you share some color on this quarter and next quarter's utilization rate?

And the second question is we noticed that our programs had been shortened from 3 hours to 2 hours in some cities, which could effectively raise the ASP.

So could you share with us some color around our pricing strategy?

And will we execute the same strategy in more cities?

Zoe, in the first place in Q1, the capacity fulfilment rate is quite well.

I think compare to last year we are low single digit higher than last year.

And considering Q2, because of adding around over 1,500 classrooms, and most of which cannot be used in the summer term, so which it will drop down a little in these ratios.

We probably can see it's about low single-digit decline compared to last year.

But with all these classrooms coming into use in the semesters in Q3, that will come back to normal.

And for the full year, we maintain our outlook in this ratio perspective.

We will continue to increase that rate by low single digits.

And the second question about the time change from 3 hours to 2.5 hours in certain cities, like Shanghai, I think that the change actually is mainly driven by the student experience.

So we continue to optimize how many times the student came to our centers and studied.

We also balance the other factors like how long my teachers can work on class.

So we also consider all of these factors to make the changes.

And this has been piloted, it has been used to Beijing around 2 to 3 years ago.

So that's not new for us.

And we will continue to make it happen in other cities, based on their current reaction and market needs.

And specifically about the pricing strategy, actually, we maintain our pricing strategies, which will take the price up in one city every 2 to 3 years.

And again, in my eyes, I strongly recommend, I strongly believe that enrollment growth to get more market share that is the most important way, fundamental way for us to grow our business.

Pricing is the last weapon we've used to do that, and in most cases we prefer to grow more enrollment.

And we don't want to take the price up that fast.

So looking forward, this year, we will take some price up in several cities, based on the plan we have made in the past few years.

And for the other places, we will wait for a time around every 2 to 3 three years to take the price up.

And we have no change in the pricing strategy, and we still maintain our pricing premier to the other counterparts in this market.

Ivy Luo, Macquarie

I have two questions.

One is regarding your summer promotion program, so wondering what's the expected retention rate for your summer promotion classes into the autumn, knowing that overall retention rate is really high at 80% to 90%, so just wondering for the summer courses, what's our expectation on that?

And my second question is regarding on the teacher front, what's our teacher salary premium over the market average?

And on our teacher selection front, what's the percent of teachers that actually gets selected to our teacher pool and the training period usually that it takes?

I think for the first question about the summer promotions, you ask exactly about the retention rates.

In the first place, I think the summer promotion we'll only do that in Beijing and a very small number of cities.

We mostly target on the first-grade junior high students.

And our normal retention rate is between 80% to 90%, while the promotion definitely will be lower than that.

And today we are delivering summer classes to the students now.

We only close, I think, one term of that.

So we still have 5 or 6 terms to go.

So today it's too early to say what kind of retention it will be.

I probably will have more color by the end of maybe next quarter's earnings call.

And what I can say is actually based on experience that we have run promotions in Beijing in the past 2 years, actually we have maintained the very healthy retention rates and a little bit better than the counterparts in the past.

So today, based on what I see, what I can get from early feedback from the business team, we are still quite confident and we still feel good about the very healthy retention for summer promotions to the fall semesters.

That is part of the reason why we can take Beijing back from almost flat enrollments to more than 40% enrollment growth in the past two years.

So I believe all our things will contribute to a much bigger and much better second half in the coming two quarters.

And the second question about the teacher premium.

Frankly speaking, we don't have statistics to show how much premium I have over the other counterparts.

And we will maintain the same strategy to give certain percentage of revenue as the teacher compensations.

And this ratio continues to decline a little bit from the 2 to 3 years ago.

I still remember 3 years ago, the teacher compensation of net revenue was around 22%, and today it's around 20.5% or 20.7%.

So we maintain our strategy to provide competitive compensations to my teachers.

Because the teacher is the very important or maybe most important investment we have in the company.

And we wouldn't change this strategy.

We continue to focus on people.

We will make sure our compensation is very competitive in this market.

And based on what I see today, we don't see we have a big number of teachers lose to the other ones, so which may be a part of evidence to show we are still very competitive in compensation perspective.

The third question about the selection rate of the teachers, I think in general 100 students come to apply in our job.

Only around 1.6% to 1.8% students can be enrolled as our teachers to be trained.

So we still maintain our very high standards to hire the best people to become our teachers.

And we don't see we will change that strategy in the short term.

We will maintain our high standard to make sure we can provide the best quality of instruction to the students.

Alex Liu from Daiwa Capital

I'm just wondering, could the management comment a little bit on the regulatory environment tutoring business, and whether the changing regulation environment has changed our capacity expansion strategy.

And follow-up on the capacity, I think we accelerated capacity growth again this quarter.

Should we think about this year's capacity growth to be more front-loaded?

In the first place, I think we have always cooperated with local governments, and we understand what the governments have done in the past, actually so we want to ensure only the qualified companies can offer their services to the students, because education is very fundamental.

And this is very important to protect the parents and students, and we as one of the players in this sector, we also feel very good about that.

And as you know, we are quite different from the other counterparts.

Actually we always take a very careful approach in entering new cities and doing new business.

So we only establish one center first, and then we ensure the center is running quite well.

And then we can open the other one.

And we, as one of the leaders in this sector, we fully agree and we will also follow the government directions to try to improve our operations and to continue to provide high quality of service to the students.

We also believe that the government directions and the government administration in this industry will help especially when the market is very fragmented.

And so we believe that the new policies of this new measures and new ways and new directions, will be beneficial for the whole sector, and will help the sector grow more healthily and sustainably.

We, as one part of the sector, we will try our best to comply with all the policies and make it work.

And your second question is about the capacity.

I think last year we grew our capacity a lot.

And this year we want to control it a little bit.

So my guidance to the team is around 30% to 50% capacity growth this year.

And we're probably right in this range.

Q1, we added 1,532 classrooms, which is less than 20%.

So we are still on the right track to adding the range of capacity by 30% to 50%.

Fan Liu, Goldman Sachs

Would you mind sharing with us the headcount of teachers and teacher assistants this quarter, and also maybe the recruiting plan for this fiscal year?

I think by the end of this quarter, in general we have around 20,000 employees.

And most of them actually are teachers.

So every year we continue to hire thousands of teachers to come to our team.

Because we still maintain standards to let people in, we still have good enough capabilities to hire and train more teachers.

We have 4 or 5 training centers across the country, and we are running them all on a central kitchen, so we have enough capabilities to do so.

And what's the second question?

I'm sorry.

I didn't follow you.

So how about the recruiting plan this year?

The recruiting plan actually we are pretty much on track.

We don't have anything special on this.

In general, I think when we do the yearly budgeting, last Q3 and Q4, we will give enrollment budgets for all the cities.

And based on those budgets, we will evaluate how many teachers we need to hire.

And then we'll run full engines to let them in.

And this year, because you know we have pilot a lot of new technology, including a new business model, like dual teacher models, actually we don't require that many teachers as what we need in the past.

So we maintain a very good pace.

Actually we grow the revenue faster, while the number of teachers growth will be lower than the revenue growth.

So we'll continue to improve our kind of efficiency in the coming few years.

And we don't have any special plans.

Everything is on track.

Han Chik, JP Morgan

I just have one question.

Did you mention that the stock split, so current 2 shares equals 1 ADS?

So the new ADS would be equal to one third of a share, and the number of ADS would go up by about 6 times.

Is that correct?

Yes, you're right.

What's the time frame for that?

The effective day will be August 16, which is subject to SEC approval and on our filings.

Alison Lee, CLSA

I was wondering if you could tell us a little bit more about the other income of USD 6.9 million, and also the revenue growth in the top tier cities for Q1.

Are they still growing in revenue?

Can I clarify the second question because the voice is a little bit down.

Could you let me know what is the revenue growth?

Yes, the revenue growth in the top tier cities for Q1, so like Beijing and Shanghai and all that.

So the first question about the other income, most of which is because we have sold a stake in one of our invested companies to the other one.

So we get kind of this disposal gain.

As you may know, I think starting from 3 years ago, having invested a lot of companies and some of them are running quite well so we acquired them to become one part of my business, like Firstleap and Shunshun.

And some other companies which are still doing well, but we figured out actually we don't have any synergy with them at all.

We have to do some in-house developments.

So we decided to sell some of the stakes to the other companies.

So this quarter, we are happy to see we have-- we sold one common stake successfully to the other companies, so which we can give us a very positive outcome.

And in the coming quarters, we will continue to evaluate all the portfolio, what we are seeing, the same as what we did this quarter, which is several capital gain from disposal compared to the number we have invested a few years ago.

So we will continue to evaluate all the portfolio companies.

And we will continue to do the evaluations and to try to measure their performance and measure their synergy with us to decide what's the next step for them.

And the second question about the revenue for top tier cities, I think, again our lowest revenue growth city was Beijing, which is over 40%.

Shanghai has recovered from last year, and grew faster than Beijing.

And Guangzhou and Shenzhen are also growing quite well, if you have heard clearly our earlier remarks.

Shenzhen is actually growing more than 100%.

So most of our cities are doing right on track, and we can foresee this very good momentum will continue in the coming quarters.

Wayne Wang, HSBC

So my question is mainly on the online course segment side.

Can management give us any more color on future growth expectation and maybe competition versus other peers, and also maybe expansion, synergy, et cetera?

I think online has great potential in the future years.

And I think part of the evidence you can see is in the past few months actually it's some online education companies who have been investing a lot by the other famous funds or famous companies there.

So this a very good sign to let us know actually online is a very good direction.

And we are the first mover in this industry.

Actually we started to do online classes starting from the year 2009.

And today, I think, last year we also transformed our online from prerecorded content to live in the K-12 area.

And this has contributed a very healthy revenue growth, more than 100% over quarters in the past 3 or 4 quarters.

And we continue to foresee this growth momentum can be sustained.

And we even see the revenue potential and the market potential for online will be much bigger and bigger.

And compared to the other counterparts in this area, we believe online is not only a simple change or a simple movement to move the students, to move the teachers and students from offline to online.

I think that represents a unique learning experience, as we have launched the new AI platform and AI technologies for Xueersi online school.

Actually we used some face recognition technologies, voice recognitions and other things to enhance the learning experience.

We also invested a lot to development new learning management systems to manage the whole learning behavior and process for all the students.

So all in all, we still believe online has great potential, and we'll be much more and more important in the future.

But again, they still need to take some time.

And we still need to invest in this area to make sure that the learning experience can be closer and even be better than offline.

So we believe the online future will be much brighter, and which will contribute much bigger, maybe a meaningful contribution next year.

Xiaoyu Yang, CICC

My first question is regarding your subject expansion plan.

Can you offer more on your plans for this year?

Like how many cities are now covered by your Chinese and English subject, and how should we expect the expansion pace in this year?

And do you have any plans for the STEAM courses?

My second question is regarding the summer promotion.

The average per student cost of enrollment and how much in the deferred revenue is contributed by enrollments of summer and autumn session courses?

In the first place, actually we are continuing to add more classes, more subjects on math that we are doing best.

And I think in the past several quarters we continue to expand the coverage of English.

Today they have covered around 10 cities on top around the 35 we have this year.

And we can foresee, we will continue to expand to the other maybe 3 or 4 cities next year.

And Chinese is also a very important subject.

Today we have around 5(corrected by company after the call) cities, and we will quickly try to expand to the other cities this year.

And the revenue growth from Chinese and English is much faster than average, and we feel very good to see that.

And for the STEAM classes, actually that is special for our Mobby business.

Starting from 2 quarters ago, we have transformed our Mobby business to focus on STEAM, which we cooperate with MIT and other famous university to develop a new technology class offerings to the students.

Today I think there the STEAM class is still a very small percentage of my revenue.

But based on the feedback from the students and the parents who have joined the STEAM classes in our learning centers, their feedback is very positive.

And we continue to expand these areas to the other big cities.

Specifically on the summer promotions impact on the deferred revenue, I think it's very minimal, because summer promotion actually is low-cost.

And in Beijing when we took promotions to the first-year junior high students, it's only 50 per subjects, so which do not give us any big revenues in deferred perspective.

And most of the deferred revenue is coming from the tuition fee we have collected from Xueersi Peiyou small class business from the students who pay us for two quarters, which is pretty much on track with my top-line revenue growth.

Lucy Yu, Bank of America Merrill Lynch

I've got two questions.

One is regarding the subjects, what's the revenue breakdown by subject, i.e.

math, English, Chinese, and others?

And secondly, is how much of the revenue is contributed by Beijing, Shanghai and also the top 5 cities?

Thank you.

I think in the subject perspective, by the end of today, most of our revenue is still coming from the math and science, I think which is more than 60% coming from math and science subjects.

And Chinese is below 10% of my revenue, and English is around 20% of my revenues.

And your second question about the revenue growth for revenue contribution from the big cities, I think Beijing and Shanghai, I think starting from a few quarters ago, we have reported is actually the revenue coming from the top 5 cities, which is Beijing, Shanghai, Guangzhou, Shenzhen and Nanjing; the reason is because actually Guangzhou is similar size and sometimes even bigger than Shanghai already.

And Shenzhen is also catching very fast.

This quarter it's more than 100% revenue growth.

So the top 5 cities actually contributed to 61% of my revenue, compared to last year as over 70%.

So we have seen the revenue growth from other cities, which is not in the top 5, is much faster than the top 5. This quarter it's around 88%.

Can you disclose Beijing as a percentage of revenue?

Sorry, we don't disclose the numbers.

But what we can say is actually Beijing's percentage of revenue is much lower than what we were 3 years ago.

Eric Qiu from CCBI

I have a question regarding the new learning center opening pace.

I was just wondering, what's your plan for new learning centers opening for the next 3 years, and separately for Peiyou, Mobby, Firstleap and one-on-one class.

Because the second question is regarding to the one-on-one class.

Because I saw the revenue percentage stayed the same on this quarter as last quarter, 13% of the revenue.

That means one-on-one still grows quite fast in terms of revenue.

What do you think is the revenue contribution, and how would you think about this business, given that counterparties are lowering their revenue contribution from this one-on-one or VIP course?

In the first place, for capacity, I strongly recommend we look into the classroom numbers.

We are quite different from our counterparts.

Actually we spend less time on the learning center numbers, but we spend more time on the classroom numbers.

Because our learnings centers will grow their size in number classrooms much faster.

So I think in classroom perspective, I think the coming 3 three years we still maintain we try to add around 30% to 50% of classrooms every year.

This will be our directions and our strategy.

So we don't change from so many years already.

And that's the right number and right pace for us to continue to be very healthy.

And your second question is about the one-on-one business.

Yes, your findings are very good.

Actually I think our one-on-one business this quarter has grown by 55% in renminbi terms, which is much faster than the other counterparts.

And the percentage of one-on-one business this quarter is 13.2%(corrected by company after the call) compared to the same year-ago period, which is 13.1%.

Actually it's slowed down.

And part of the reason why the one-on-one can still grow much faster that is because we have improved the alignments between the Xueersi Peiyou and the one-on-one.

This year we have more than half of my one-on-one students coming from Xueersi Peiyou business.

So we continue to see we have capacity to transform more and more needs to the one-on-one, which will significantly reduce the new customer acquisition costs.

So every day when we talk about the VIP business, actually the most challenging for them is because they spend more and more money on one new customer.

And so if we can successfully maintain to convert the students from Xueersi Peiyou to one-on-one and fit the needs, we are happy to see we can continue to improve in this area.

So I personally still treat the one-on-ones in the complementary services to Xueersi Peiyou.

And we will make sure both segments can grow very healthy in the coming few years.

Thank you.

That's all we have for questions.

And with that, we conclude our conference for today.

Thank you for participating.

You may now disconnect.