Ladies and gentlemen, thank you for standing by, and welcome to the TAL Education Group Second Quarter Fiscal Year 2018 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 you that this conference is being recorded today, Thursday, 26 of October, 2017. And I'd like to hand the conference over to your speaker today, Ms. Mei Li. Thank you. Please go ahead.

Thank you all for joining us today for TAL Education Group's second fiscal quarter 2018 earnings conference call. The second 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 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 strong ongoing growth momentum in the second fiscal quarter 2018 results. Our top line growth continued to be driven by high demand in all cities and supported by further capacity expansion. Top line growth in the quarter was 68% year-on-year in dollar terms to US$456 million. In RMB terms, net revenue grew by 70% year-over-year. Enrollments increased by 101% year-over-year. GAAP income from operations increased by 32.6% to US$68.3 million in the second quarter and non-GAAP income from operations increased by 33.7% to US$79.9 million. The decrease of operating margin was mainly caused by seasonal expansions of new teachers in classroom capacity. I am sure as the slow ramp up of overseas [consulting business] will have occurred last year. In the first two quarters of fiscal year 2018 we have added 2,021 Peiyou Small Class classrooms. Total number of Peiyou Small Class classrooms increased by 74% year-over-year, compared to the number of August 31, 2016. The new classrooms added so we have new teachers hiring that spring have not yet fully used and we will bear a cost of this new capacity expansions in the second quarter. The combined teachers and renewal costs as a percentage of revenue is higher than Q2 last year, and these will become more normalized as the new teacher continue to take more classes and the new classrooms become fully used in the second half of the fiscal year. Now Mei will keep on updating our operational progress in the second quarter. After that, I will update you on our ongoing investments in technology, which is guided by our long-term strategy revisions of advance in technology -- advance in education systems and technology.

Our revenue growth was supported by the growth across all business lines in all cities. Small Class which consists of Xueersi Peiyou Small Class, Firstleap, Mobby and some other educational programs and services accounted for 80.9% of total net revenue compared to 85.1% in the second quarter last year. Peiyou Small Class which remains our core business represented 73% of the total revenue compared to 77% in the same year ago period. This lower revenue contribution from Xueersi Peiyou Small Class was mainly due to the faster growth of online course business and the consolidation of the newly acquired businesses. Net revenue for Small Class was up by 58% in dollar terms and a 62% in renminbi terms, while the enrollment increased by 77%. Xueersi Peiyou Small Class revenue generated from cities other than top 5; Beijing, Shanghai, Guangzhou, Shenzhen, Nanjing, grew by approximately 70% in renminbi terms. Cities other than the top 5 accounted for 42% of the Peiyou Small Class business, up from around 48% in the same quarter last year. We have achieved a triple digit year-over-year renminbi revenue growth in terms of the 25 cities that we have entered by the second quarter of fiscal year 2017, including Jinan, Shijiazhuang, Changsha, Qingdao, Luoyang, Nanjing, Ningbo, Hefei, Wuxi and Fuzhou. The revenue growth across all cities was driven by incremental ramp up in enrollment from our early classroom expansion. Let me give you an update of the summer promotions. We offered summer promotions in Beijing and a limited number of cities given the retention rate into the short term of these promoted courses, we expect the enrollment growth in the coming quarters to continue to grow robustly. For our one-on-one business we had a good second quarter and achieved a year-over-year revenue growth of 96% in renminbi…

Thank you. I would like to give you an update on our progress on the technology investment. We lately announced the set up of joint research center with Tsinghua University. The joint research center will work on applications of smart technologies into each segment of our facilities. The technologies including computing vision technology, speech recognition technology and natural language processing technology [architecture] will be applied in the real teaching scenarios on TAL's [indiscernible] today. And help us to give the students a more efficient and more effective and more choice experience. We believe mainly narrative applications of smart technologies in education context will be noticed and generative from many education companies. Our exposure -- our aspirations of education plus AI is still very early and ground breaking stage of what will be called revolutionary new models of learning. Let me now move to our outlook for the next quarter. Based on our current estimates total net revenue for the quarter of fiscal year 2018 is expected to be between US$411.7 million and US$416.9 million representing an increase of 58% to 60% on a year-over-year basis. This estimate reflects our current expectations which started to change. That concludes our prepared remarks. Operator we're now ready to take questions.

Thank you, sir. [Operator Instructions] Our first question is coming from the line of Alvin Jiang from Deutsche Bank. Please ask your question.

Hi management, thank you for taking my questions. I have two questions. The first one is on margins. This quarter's margin is around 450 bps decline year-on-year. What are the exact underlying reasons for this margin decline and what's the margin outlook for maybe next two quarters and also long-term margin? Thank you, I have a follow-up.

Thank you, Alvin. I think your question is very important, which -- or just so many people has questions about this. To give you kind of the details of Q2 why the margins declined, we have three very good reasons. The first one is as we have mentioned in our prepared remarks, we have continued to add more capacities in Q2, I think Q1 we added over 60 learning centers, for our Xueersi Peiyou Small Class and Q2 is 20 plus. And all of these new classrooms and new teachers we hire correspondingly, we will gradually trying to fully use in the coming maybe one or two quarters. In general we spend around 12 months to have let me say it, reach a certain level of the [peripheries] as well as in other cities, so which takes some time. But we are very good to see that with 74% capacity expansions in the first two quarters and we [indiscernible] Q3 and Q4 which can give us very solid growth foundations for next year. And our retention rate for this quarter because of the new added capacities is gradually going down a little bit by around 1.5 points and we are also quite confident to see this number will recover in Q3 and Q4. And the margins for Q3 and Q4 in [indiscernible] what we can say is if we combine Q3 and Q4 together second half the margin density in year-over-year perspective will better than Q2 we see some positive news over there. In the long run we still believe the company will on the right track and we will try to pace our CapEx expansions and try to fulfill our classrooms based on what they need, and will try to manage how it grows. So we don't hurry to do something to put a number on, we just follow our own way and our pace to manage the whole business. So, we are going to see some recovery in Q3 and Q4.

Got it, thank you. The second question is on the growth, we notice that the deferred revenue growth slow down a bit in this quarter and does this make any impact to the growth of next quarter and how should we look at the full year revenue growth and long-term revenue growth? Thank you.

Thank you, Alvin. I think I didn't answer the first question very fully. I think the -- so let me give more colors about that. The first I think we declined margin by 4.5%; well we have the leverage [indiscernible] margin by around 5.1 points. I think 2 points of that is because the capacity we have added in the first two quarters which as I have mentioned just now. We also have around 1 to 1.5 points there because of promotion. The promotions include two things. The first one, we are running the RMB50 promotions in Beijing and other maybe two to three cities and only for the new children for the first year junior high and the first year senior high. So of all these promotions actually give us a very good return and we can see it in Beijing, we have seen where how it grows over 40%. And this can also give us very good foundations when moving to second half of this year. And we also running some promotions for online, which is for the Xueersi online school, Xueersi online sale. Xueersi online school is also running the RMB1 promotions, which is also give us a very good return, which can be proved by the top line revenue growth for online growth by 144%, while the [indiscernible] for online school by -- very close to 200%. And the other 2% actually the 2 points is based because newly acquired business starting from last year Q3. Q2 in general that is a bit positive for them, because most of the students they register offers in Q2. So part of the reason is because of the unique nature of the overseas consultancy business, actually, I think the first one or two years they are in loss making…

Our next question is coming from the line of Ivy Luo from Macquarie. Please ask your question.

I have two questions here as well. So one is, could you elaborate more on the summer promotion in terms like how many enrollment do we get from those like discounted from a promotion cost, and actually what was the number last year and what's the retention rate for those students? I have a follow up, thank you.

Okay, Ivy. I think for the promotions, for offline promotions that's for just Peiyou I think mostly would depend in Beijing, pertinent to last year we ramped promotions for their students in the first grade junior high and the first grade senior high and the enrollments for Beijing in promotion studies goes quite well and the retention rates are also better than last year's over 50%. And that's part of the reason why we see Beijing's performance continuing to grow over 40% in past three quarters and this growth will continue you know a little bit [encouraging] in the coming one or two quarters. For the online the Xueersi online school promotion that's first year we did that and the retention rates map is lower than the offline school retention rate because that's online, but also it's better than what we expected. So I think as I have mentioned in Q2 because of our these promotions we grow the Xueersi online school enrollments by close to 200% while the top line revenue is more than 144%, so we still think for online that is the right way to do that. So that's pretty much for the summer promotions. Compared to our counterparts because we're only running the promotion in world's largest cities as well as the grades so we don't run and we don't have any intention to run the national wide promotions maybe even next year. So we'll still evaluate and pilot the promotions very carefully. What we [tell more] is we need to make sure the quality to the students are still well deserved, and we need to -- and we'll be very careful to use promotions more than where we tend to today.

Thank you. And second question on the expansion, I just want to clarify that, how many classrooms did we add this quarter, sorry I didn't get the numbers clearly? It seems that there's a bit slowdown from 1Q is that seasonality impact and because we say we're going to pace expansion a bit in 4Q and 3Q. So I just want to know like, what should we expect in terms of classroom expansions in second half of 2018?

Let me clarify the numbers. For the first two quarters, we have added a classroom numbers which is around 2,000 classrooms, while specifically for Q2 we've added 489 Peiyou Small Classrooms, 489. So compared with Q1 yes we slowed down a little bit, Q1 we added around over 60 learning centers, Q2 is only around 20 plus and Q3 is similar level around 20 plus.

Our next question is coming from the line of Leon Chik from JPMorgan, please ask your question.

Hi, just one question, your G&A cost was up 42% year-on-year and your sales was up 68%, why is it going up slower?

Thank you, Leon. The overall SG&A growth over 60% which is below our revenue growth, part of the reason why we have the higher selling and marketing expenses gross, that's because in the summer term that is a very important time for the students. So we start to run some online marketing for Xueersi online school -- Xueersi online school specifically. So this kind of online marketing spend will record into our Q2 telemarketing. And based on our checking of the online actually with -- we believe there's still a good model to grow but we will only do that in the summer term because that's the work off season and big season for the students. For Q3 and Q4 we don't have similar plans.

And G&A where is that down, I mean where is not growing as fast as sales? Thanks.

Yes. For G&A that is because you know most of the G&A -- actually that is because our very own financials people, we as a company even with growing our top line we are still want to make sure our financials will grow much lower than the revenue growth, and we relatively do a good job. Here in the past few years we don't increase our number that much in the G&A expense [indiscernible]. All of the investments we will put them into the IT, technology and product costs. So we control very carefully our very own financials that's part of the reason why. I think starting from two to three or four quarters ago our G&A has grow much lower than our top line of revenue gross.

Our next question is coming from the line of Fan Liu from Goldman Sachs. Please ask your question. Q – Fan Liu: Thank you for taking my questions. So my question is that one quick question is actually just want some color on what's your enrollment contribution for ROI? And also including -- if we including the summer promotion what will be your enrollment growth and also what will be the absolute growth a bit like if we -- excluding the summer promotion? And the second question is that, so what's -- I think that your bet revenue portion this quarter already slow down, I will take that. So what's the exact reason behind that? Anything related to regulation or they need to [indiscernible]. Just wanted to get more colors on that. Thank you very much.

Actually your first question about enrollment growth, online enrollment is around 30% of our total enrollments which is phenomenon important to us because we are gaining more market share sooner on the technology. Online enrollment growth is close to 200%, even we take all the promotions still grow still grow more than 100% which can be proved by our top line revenue for online has grown 144%. So because the promotion enrollments that's [indiscernible] revenues, so even I take them all I still grow more than my top line revenue gross. I think for online the key is to grow more enrollments, get more market share and try to establish our top tier leaders in online market, I think that's the key. Market share is the key. For expansions I think we don't have any specific reasons related to regulation or some other reasons. I think Q1 we have added 1,600 classrooms, Q2 is total 500 classrooms. Compared to where we have been in the past few years, I think 500 to 600 classes on average for [online] here actually it's quite normal, for one quarter it is quite normal. So when we have added 1,600 classrooms in Q1, of course, we need some time to digest and some time to make sure our classrooms can be fully used in the coming two, three quarters. So that is our pace of operation, we still try to manage how it goes. There is no time of specific things about that. I think we are in a free market, so the thing we are in the same environment, so we don't have anything unique or special from the other one. Q – Fan Liu: And just a one follow-up. So what's your revenue growth for Shanghai this quarter?

Shanghai is over 70%.

Our next question is coming from the line of Natalie Wu from CICC. Please ask your question.

So a couple of questions here. First one is about the revenue growth. So excluding firstly about Xueersi, just wondering what the year-on-year growth for this quarter and for next quarter's guidance for the rest of your business? Second one is about the margins. So what do you expect positive effect on margins from your teacher model to be noticed on the consolidated enterprise level? And last one is about the capacity expansion plan. So can you remind us about the capacity expansion plan for the rest of basically 2018 and also next year for more classes and to teacher model classes, respectively? Thank you.

This is three questions, okay. The first one, I think the revenue guidance or the revenue growth for Q2, we're taking all the firstly in the Xueersi. Actually, in general, we don't give this kind of guidance. So like I say in Q2, I think combined Xueersi and firstly and some others well coming [indiscernible] which is around 7% to 10% of our total revenue for Q2, in revenue mix that is. And especially for Q3, I think that is not easier to do that, because actually firstly we have acquired around, I think one or two years ago so they have been the base, especially in Xueersi, because we acquired in last year Q3. So in Q1 and Q2, they don't have the year-over-year comparisons, which is around 5% after [indiscernible]. Well in Q3, because, I think is, because they have their year-over-year comparisons and the gross rate is below average, so that's kind of [check] over there, [indiscernible]. And your second question about the margin for the teacher models. We can see it's actually -- we continue to pace our expansions on the teacher models. In the first quarter, we have entered five new cities all of these are five teacher centers. And we also try to add more teacher centers in the current cities. I think this is we are pace. The contribution from the teacher models, this year still minimal. I think the group tend to talk about we will be in the time of next year summer, because in Q3 and Q4 we are continuing to add more learning centers -- add teacher model centers. So we strongly believe next year's summer will be a good time to talk about that. And the capacity expansions for this year, I think the first two quarters 74%, the full year our guidance is around -- our guidance for the expense -- for the capacity enters will be let's say for the coming three years will be in the range around 30% to 50%. I think this year is definitely close to 50% maybe a little bit up over there and next year, I think the next two to three years we will still be in the range between 30% and 50%. So that's the case, we don't -- because we don't change our pace too much. In the future we will still make sure we can operate stably and manage it how it goes in all our business [perspectives]. Thank you, Natalie.

Our next question is coming from the line of Alex Liu from Daiwa. Please ask your question.

My first question is actually on the enrollment, so this quarter we see a Small Class enrollment growth actually accelerate comparing to previous quarters, but we actually see the capacity growth rate slow down a little bit. Just how should we think about enrollment in the second half this year and also next year? I have a follow-up.

Okay. I think the enrollment growth for Xueersi Peiyou Small Class actually is quite stable in the past few quarters, they don't change that much. And the enrollment growth for firstly with the Peiyou small classes also much better than before, I think right before acquisitions compared to that time this growth rate is much better than that. And the growth rate for the online I've mentioned it just now, summer is a kind of very good numbers close to 200% and I think the longer term prospect is we still wish our online can continue to grow maybe close to 100%, which is the right number to kind of expect. And Xueersi -- the Zhikang the one-on-one business I think the enrollment is not a good number so we don't really talk about that. So in general I think we will still maintain our healthy growth in enrollments and we more likely to see our enrollment growth will be a little bit faster than revenue growth I think in the coming quarters and we are most likely to see our online continue to grow fastest across all segments. And because of adding all of these new classrooms, we have better teachers here, I think certainly may be the winter or maybe spring we can see more and more contributions over there.

Thanks. And also would the management share with us the thinking process on deciding the timing of capacity acceleration and also capacity acceleration perhaps in the angle of competition and also local demand?

Yes. I think in general when we have tried to decided to add more capacity that is driven by numbers. We have all the numbers included to fulfill areas including retentions, refunds all of that. And we need to also carefully evaluate the parent satisfactions on this. And so when we have reached some balance of all these numbers we decide to add more capacity. There are two reasons I think it applies, the first reason is we received a lot of consumers' compliant about they can't enroll into our school. So that's part of the reason we decided to -- we maintain our quality, we have decided to spear a little bit in a capacity expense of studies. And on the other side is because we also received other feedbacks from the parents, they care more about their kid's health, so we leave more time for the classrooms to -- because of the environment protection reasons. So in general I think our the capacities -- the classrooms we are adding this year, generated pace as in two to three months more than what we did in the past and to put them into use. And we believe that's the right pace for a company to do and moving forward we will continue to bear this message to make sure we can drive the right balance between the revenue, profit and between long-term satisfactions from the parents.

Our next question is coming from the line of Lucy Yu from Bank of America. Please ask your question. Q – Lucy Yu: I have got one question. When you mentioned the 12 month in your previous speech, was that the 12 month that take a learning center to ramp up to a mature learning center or a 12 month to reach P&L breakeven?

To mature learning centers at the same level of capacity for [indiscernible] kind of the mature centers we have seen, yes it is. Q – Lucy Yu: So how long does it take to breakeven on a P&L basis?

That's depending on different cities, but in general it's around six months. Q – Lucy Yu: Six months, okay. I have one more question. I noticed that you have closed down some one-on-one learning centers this quarter. But one-on-one is still growing very healthy at about 100% this quarter, so I am wondering, what's the reason behind that? Thanks.

I think the one-on-one today we have actually have two parts. The first one is actually company [business], the second one is the overseas consulting and insurance business we put them in one-on-one. In Q2 you will see close to 100% that part of it is class extension, so [indiscernible] for the last year in July in Q2 now they have the numbers in Q2. But even we pay the overseas consultancy out the Zhikang one-on-one still grossing very well. Compared one or two years ago they are recovering so much. Today we are seeing more than half of students coming Xueersi Peiyou and which significantly reduced our new customer costs. And we can see there are no lumpiness because there is a line that is cost related to our Small Class business. With Small Classes scoring quite well one-on-one can also grow quite well. And in profitability studies one-on-one is also much better than what we have seen in the previous years. We also try to offer some group, small group classes, one teacher works maybe eight students or six students as a group. So I think we have a lot of new changes in the one-on-one business which is doing quite this year. But I can say they are doing quite well this year, so they will continue to being better and better in the future, where one-on-one still is a business we need to do work about that and we will work together with the team to make sure they can [indiscernible] growth in one-on-one. Q – Lucy Yu: So its sounds like one-on-one is improving, why are we closing down the new centers?

Some -- part of the reason why they are improving is because we shutdown the underperforming centers and we increased the size of the good performance learning centers. We centralized a little bit in the city to make sure, we can provide high classes to the students. So we don't want to really energize one business, you need to have two ways push and prove. You cannot only just give them incentives, but you also need to try to close the underperformance centers. And also [cost base] because of the rental issues, because we maybe we have running of our rentals, so we have to change their locations, so some reasons for that.

Our next question is coming from the line of Mark Li from Citi. Please ask your question.

I want to know about for the Small Class, actually for the ASP. If we exclude summer class promotion, what is the trend for the ASP for the quarter? And also what is the enrollment growth roughly, if we exclude some other class promotion? Thanks.

I'll answer second question first about the enrollments. We exclude the summer promotions for the offline business. I think it's pretty much similar, because we only run promotions in Beijing and several big cities [indiscernible]. So they don't any difference over there. And your first question about ASP, you're taking out the promotions. I think you -- we are taking out promotions and especially for Q2 the Small Class ASP has declined a little bit. Two reasons for that, the first one is, I think last year, we increased the price in the big cities including Beijing, Shanghai, Guangzhou and Shenzhen. This year, we also take price earnings in similar number of cities, but because the volume of these cities compared to Beijing, Shanghai and Shenzhen is much lower. So that part you see -- in that studies that's a little [deduct] in ASP. And second reason is actually in the past few quarters we continue to see our revenue and enrollment growth from the cities other than the top five, is grow faster than top five. So this --because of the mix reasons and they can also can see slightly the ASP were going down a little bit this year. But moving forward to next year, the future years, we still foresee our ASP can grow by single-digit, because we are also rebalancing our pricing strategy and we will try to make -- and we also very happy to see the recovery growth from Beijing, Shanghai and some other cities. So in the long haul we still foresee that low-single-digit growth rate in -- for the ASP for Small Class.

I also want to ask that, how is the economics in the learning center for our data, the total cities of Beijing and Shanghai and Shenzhen versus the lower tier cities say the margin or the enrollment per center, et cetera? Thanks.

I think for -- I think that's different by study. I think for the maths, I think across the country are quite similar. But in big cities like Beijing, we have a lot of studies and we have Chinese, we have English. And Chinese-English is growing much faster than maths. While -- when we are going into the lower tier cities today you can still see out of the new learning center today we only offer one studies. So if we combine all the things together all the studies together, Beijing's centers may be a little bit lower than the other cities but that is because driven by different studies.

Our next question is coming from the line of Alison Lee of CLSA. Please ask your question.

Hi, thank you for taking my questions. I'm asking on behalf of Mariana. I just want to know, what is the utilization rate in Q2? And also I wanted to quickly recap of the ASP for the online segment I just missed that number? Thank you.

Okay. Same as before, we don't disclose the numbers of our capacity utilization [indiscernible] anyway. But what I can say is you know year-over-year comparison study of Q2 our capacity fulfillment rates actually compared to last year's same quarter is declined by 1.5 points around. And your same question about my online ASP, my online ASP compared to last year same quarter I think is going down by around 14% because we running promotions, if you're taking out promotions actually the ASP is slightly up.

Our next question is coming from the line of Tian Hou from T.H. Capital. Please ask your question.

I have two questions one is related to the course offerings, so in different cities, certainly when you ramp up you start from some classes and then eventually expanding to other subjects. So among all your courses or classes, what type of classes are most well taken classes by the students? So in the price wise are those in the classes have relatively higher price than others and also what's the pricing strategy on a new basis? So that's number one question. Number two, recently we have several education company IPOs, so I wonder, if the IPO is going to change some in offline tutoring market landscape? That's two questions.

Thank you, Tian. You pick a very good point. It is actually every time when we go toward new cities we only start to from maths, only one subject, one single grade first, and then when we have making work and better in performance and then we will try to expand this maths to more grades. And when we have I think three or four years when we -- in the maths perspective in that local city would become number one or would become the top tier, we will decide to add new subjects maybe physics or chemistry, or maybe Chinese or English. So we'll careful to pace the new city's growth rate year-over-year. And even today we cover all of these over 30 cities but more than half of them they don't have English, more than 20 of them they don't have Chinese, so we are still pace low rate to adding more studies and more grades to the cities, based on their lifecycle. And we also see -- work carefully on their KPIs. So that's about pricing and we don't have any price difference. In general we don't have any price difference among different studies. They are all the same price for all studies. I think that it's single very effective strategy on pricing perspective across different studies. Your second question about the education company IPO, I think in the first place we as one player I think in the market well, we're happy to see more and more other counterparts, more and more of our friends they can go IPO in the public markets. Their IPO means the whole market actually is well recognizing the potential of education companies, the potentials of this market and we are seeing a lot of statistics. We can see in the…

Our next question is coming from the line of Thomas Chong from Credit Suisse. Please ask your question.

Hi, thanks for taking my questions. I have two quick questions, given our strong revenue growth in the second quarter we just pretty much in line of consensus. How should we think about a full year revenue growth and how should we be thinking about the full year margin compared to last year? Thanks.

Thank you, Thomas. I think the first question about the full year revenue guidance, I think the first two quarters we grew more than 70%, Q3 our guidance is as I indicated at 60%. And I think the full year pretty much should be in a number between these two numbers. On our guidance for next year, I mean maybe next two to three years, as I answered it before is in the range between 30% to 50%, which is no change and we are quite confident to deliver that in two to three-year timeframe. And the full year margin as I said in Q1, we are on 1 point decline in [indiscernible] Q2 is 4.5 points. And second half year-over-year comparisons will be little recoveries on Q2 and so that's the case. As in for full year numbers we probably can give you much of the numbers next quarter.

Our next question is coming from the line of Eric Qiu from CCBI. Please ask your question.

My question is regarding to what's Beijing's revenue percentage and also what's the top five cities revenue percentage to this quarter and also in first quarter, because I know this last year, it's about 30%? And then my follow-up is, you said, you basically launch the summer promotion in Beijing. So I just want and what your intention and why that you didn't choose to launch the promotion in all of the cities you have entered? Thank you.

Thank you. I'd say the top five cities revenue contribution was said that in the previous remarks is around 58% that's for top five, which is 2 points lower than the same quarter last year. And if we put them in the longer term or a two year timeframe, I think this number is much lower than before. And specifically around Beijing, Beijing's revenue contribution is below 20%. And I still remember we have first [indiscernible] three years ago the Beijing contribution is more than 40% that's the number over there, today it is below 20%. And your second question about why we don't choose to replicate promotions across the country, that's [a departure]. I try to answer in macro and micro perspective both. In the macro perspective, we need to be very careful you know, promotion sometimes is very deceptive in the short-term. Sometimes you look -- you take quite weird because you know we have lowered down our price so much that if you're happy about that, so they will come for you. But promotion sometimes dilute your focus on the quality and promotion sometimes will be, if you do something too easy, you will forget something tough. But only when you want to do a lot of things more difficult and more tough you can build out your core competencies. So we still believe in the interest of the parents, in their studies what they care more is when they decided to send their students to your school and when they graduate from your school, they have making progress in their performance. So the quality is the first priority and the quality is the most important thing they will care about there. We also tried some promotions before, but we quickly try to control the level of…

Ladies and gentlemen that does conclude our conference for today. Thank you for participating. You may all disconnect.