Ladies and gentlemen, thank you for standing by, and welcome to TAL Education Group Fourth Fiscal Quarter and Fiscal Year 2017 Earnings Conference Call. [Operator Instructions] I must advise you that this conference is being recorded today, Thursday, 27th of April, 2017. I would now like to hand the conference over to your speaker host today, Ms. Mei Li. Thank you, ma'am. Please go ahead.

Thank you all for joining us today for TAL Education Group's Fourth Fiscal Quarter and Fiscal Year 2017 Earnings Conference Call. The fourth 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 Management 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 those 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. 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 set of fourth quarter and fiscal year 2017 results. We continue to enjoy robust top line growth, driven by the high demand in all cities and supported by further capacity expansion. Top line growth in the quarter was 80.7% to USD 316.3 million. In RMB terms, net revenue grew by 90.7% year-over-year. Two key factors I would like to make about our top line performance and draw your attention. First, revenue growth was supported by our core small class business across all cities, with actual growth lead from a classroom capacity that we have added in the previous quarters. And because of the earlier timing of the Spring Festival holiday, we scheduled one more section of spring classes in the fourth quarter in most cities than we did in the same period last year, which is definitely better. And second, the contribution from the newly acquired business, Firstleap, in particular, which didn't contribute much last year but grew healthily and contributed more revenue than expected in the fourth quarter. For the full year, net revenue was up by 80 -- was around 68.3% in U.S. dollars and up by 78% in renminbi. Non-GAAP operating profit increased by 113% year-on-year, also above our key expectations. Earlier in the year, we started to accumulate classroom capacity for the summertime, so the non-GAAP operating margin declined in the first quarter and the second quarter, but gradually recovered in the third quarter and fourth quarter. As a result, for the full fiscal year 2017, non-GAAP operating margin was 16.4%, down by 150 basis points from prior year, declined slighter than our early expectation. Today, I would first…

[Operator Instructions] Your first question comes from the line of Zoe Zhao from Crédit Suisse.

I have 2 questions. First one is on your margin guidance. If you could give us some colors on both GPM and OPM, that would be great. And the second question is regarding your summer promotion strategy. If you could talk a bit about like the scope of the summer promotion and what's the pricing you are going to set, that would be very helpful.

Thank you, Zoe. The first question on the margin guidance, same as before, we don't give any very detailed guidance on margin status. But in the direction status, we probably can say I've been looking forward to our fiscal year 2018, where is that, most likely, our margin will be flattish considering we have some recovery from the newly acquired business, especially the Shunshun. Firstleap, we're probably going to see more positive outcome from them. But at the same time, we need to also balance because we are still an economic-driven company. So we need to invest for some of the new business, maybe for -- through a 5 year's time. So today, we're actually at the beginning of the year. So my current view is we probably will be flattish in margin status. And specifically about gross margin and OP margin, we don't have that much details to disclose. But again, we're [ as calmly ] very cautious and very serious about our margin, so we will balance our investments and our -- the kinds of the leverages we can get in margin status. So we will deliver a sustainable and reliable growth in the future. And your second question about the summer promotion. Same as before, we don't have any national-wide promotion plan, and we don't see any intention we want to increase a lot in promotion status. Promotion is one of the tools we could use, but actually, we still strongly believe that teaching quality is the final say in front of the customers. Compared to promotions, we would like to invest more to -- on our teacher, on our teaching system, on our learning management system to make sure that the whole experience that we deliver to our students are the best in the market. So in direction status, we don't have any much bigger promotion plans. We'd probably do something similar as already last year, especially, most of promotions will happen in Beijing. Some of you guys have seen our [ ads put out ] already. So again, we treat the quality as the most important stuff we need to deliver to the children and we are very serious about that. And if we have anything more to update, I will let you know.

Our next question comes from the line of Terry Chen from HSBC.

I also have a question on margin. So I guess the -- one of the positive color in this quarter's results is that your non-GAAP operating margin expanded by 2.6% year-over-year. You just mentioned that you're guiding for flattish margin. Maybe it will constantly progress. So I just want to know what's the reason behind the margin expansion in this quarter, and why do you think it won't extend into FY '18?

Okay. In the first place, as in Q4, our margin is a little bit higher, and why is that? That is also because we have one more [ sections ] because of the timing of Chinese New Year is much earlier this year. So the one more section, putting into 1 quarter, we only have 13 weeks. So the one session actually is quite meaningful for both our top line revenue and the margin -- and the profit status. And if we go back to see our full fiscal year 2017, we have the margin decline in Q1 and Q2 because we tried to expand more capacity at that time and we have some more new teachers. And Q3 and Q4, we meet the capacity we are going to use, so we have a little bit coming back in the margin status. Looking to full fiscal year 2018, in the first place, we are happy to see the capacity we have amassed in the past few quarters. Now they have contributed more and more. We -- also, our key, a very healthy pace to try to expand more this year. What we want is healthy, controllable and sustainable growth. Healthy, controllable and sustainable growth. And on the other side is we're still focused a lot on technology. Our mission has just been clarified, is we need and want an education through science and technology. With this special mission, we definitely need to invest in our data and technology platform to try to give some new experience to the students. Saying that some of the new pilots we're running in our one-on-one centers and in our small class centers, where we're trying to present a new technology to make sure the experience and the quality we deliver to the students are much better than before. So this is a right balance between -- we have some buffers, we have some leverages we can use in margin status, but at the same time, we need to invest to [indiscernible] . So coming to today, we believe our guidance is kind of -- is too early to give, but the direction status will try to make sure the margin will be flat. But flat doesn't need every quarter to be flat. Maybe in some quarters, it's a little bit down, in some quarters, it's a little bit higher, same as what we see in last year. So that probably that's the key driver for our margin status.

Okay. Yes. Got it. Just a quick follow-up. Can you remind us about your capacity expansion target for FY '18?

Our capacity expansion target, I think that comes in 3 years, so we'll still be around 30% to 50%, and 2018, because last year, we just added more than 70%. So this year should be close to the high end. And again, I think the capacity added is not driven only by the market demand. It's more driven by our quality controls. We don't have any intention to say I just want to grow my revenues so I just add whatever we want to do. We still set a series of KPI and metrics to control the pace of CapEx expansions. So we will even be more cautious and serious about them this year because we need to make sure our -- the capacity expansion pace is aligned with our new teacher recruitment and our new developing technology. So the teaching quality is the highest, our first priority in our company. So we will be paced, and our goal is, listen, let's try to deliver very healthy, controllable and sustainable growth.

Your next question comes from the line of Alex Liu from Daiwa.

I'm just wondering, I think the management just mentioned the positive tailwinds from the favorable demographic trend in large cities in China. I'm just wondering how long does the management think this demographic trend will last, especially in cities like Beijing, where we already found a very sizable market share.

Okay, Alex. And I think this trend should be proved by the statistics in major cities in China in the past few years, new -- mainly new populations. Such a thing in Beijing, we have seen this trend as going up since the year 2008. And even coming into the most recent year, the year 2015, the number is even higher. So because we are K-12 providers, when one student is coming into my system, it takes them a very long time to stay with us. So it's difficult for me to say how many years we can enjoy this kind of a favorable trend. But I think as the industry studies, this is really a very important driver for us. And based on the statistics we have seen, at least, until we see the new baby rate start to decline in that year. So we, probably, we still have a long time. We could be favorable by the new baby trends. And last year, we even see, with the end of the one-child policy, more and more families in the mid-cities, they start to have a second child. And this is also a very important driver in a set of studies. This trend is not only beneficial to us, it's beneficial to the sector, and it is also beneficial to the whole industry. So we, as a company, we -- all we need to do is not only about this trend, but what we need to care is whether we control or deserve to be such successes. We need to be very careful about the pace of our new revenue growth in the coming years. And we still need to improve quality as a first priority, and we still need to manage a healthy growth, which is more sustainable and which is also more beneficial to our shareholders in the long run.

Our next question comes from the line of Fan Liu from Goldman Sachs.

I think Rong Luo has mentioned before 40 cities might be the cap for the offline network. Do you still hold this opinion now? Another related question is, do you see any challenges to your management [indiscernible] wise during the network expansion? If yes, what's your strategy to solve it?

Yes, Liu Fan. I think previously, we believed if we only use the offline model, maybe we only penetrate around top 40 -- maybe a little bit more in the whole China market. But in the past few years, I'm seeing ways that to invest to online technology, including the pilots of running [indiscernible] platform and including software and new technology we have to use in Xueersi, Guangzhou and some other places. So -- and the double-teacher model is also very important on various successful pilots we're going to run. Even coming to today, we can't say the online model, the double-teacher model is fully successful, but we still believe they have great potential. So with all of these advantages coming from technology and the Internet, so we probably may have more opportunities to leverage this new technology and Internet and try to penetrate more. But again, today we are still in the very beginning of this new wave of development. And currently, we only cover 30 cities. And in these 30 cities, a lot of them, they have the -- I'd say more than half of them, they have more than -- less than 10 learning centers. So what we need to care today is not to count how many cities we can penetrate. What we need to care is what we can do more to make sure in current cities we are penetrating, we are quite -- we are still quite successful. In current places we have presence, we collaborate their [indiscernible] technology to try to cover more students and give much better level of quality to the students. So we need to care at current market first, make sure what we deliver is the best in the market. So that's our key priority. And how many cities we can cover? How many places we can go? I think that is the natural results if we drive everything right in the beginning. So coming to today, I don't have any hesitations to say, hey "How many places we can cover?" But we believe if we continue to put our students in the first priority, continue to improve our teaching quality, continue to invest in data and technology to leverage the power, so we'll face a much longer growth in the long run.

Our next question comes from the line of Tian Hou from T.H. Capital.

Two questions. One is regarding your pricing policy. We did see some price increase for some courses and even in some cities. So I wonder, what's going to be the pricing policy for the year? And what could be the impact to the utilization rate in those cities? That's the first question. Second, Shanghai initiated a policy which gave the students some in-school, like, training by the school themselves. And I wonder, what could be the potential impact to you guys? And what do you see this policy, the impact nationwide? Will this policy go on to the other cities in the future? What's your view on that? That's my 2 questions.

Thank you, Tian. In the first place, our pricing policy, out of [indiscernible] , relating in the past, we would take price hike in one city, average 2 to 3 years. Last year, we did price hike around 8 cities, and this year, probably the other 6 to 7 cities. And we don't see any direct connection between a price increase to any capacity fulfillments. And our capacity fulfillment actually is driven by how many new classrooms we can provide to the market. And so far, based on what we see, actually the rate is quite stable. And your second question about the Shanghai policy. I think, again, we as a K-12 tutoring company, we worries more percentage of this market share, and in the whole week, we -- the students only spend few hours in my classrooms. So we are supplementary facilities to the mainstream facilities. So we call us tutoring, not education. So in this perspective, what we can see is we don't see any direct impact from them, and we're also watching very closely about the other policies -- or maybe new policies possible to happen in other cities. And again, we are only a tutoring company. We are supplementary to the school. So, so far, we don't see that much impact from there.

Our next question comes from the line of Mariana Kou from CLSA.

Again, congrats on the good set of results. My -- I have 2 questions. My first question is on the margins for your other business segments. For example, [indiscernible] mentioned but maybe you could share a bit of color on how should we think about the margin across the segments? It would be very helpful. And my second question is bit on technicality on the ASP flow. Should we expect a bit more flow in Q1 and Q2 just because there will be a bit more uplift from Firstleap?

I think -- in the first place, because we manage our segment, we only have one single reporting segments, nothing else. So in general, we don't disclose that much details about the margins by different segments, because we don't manage in that way. But in the rest of study, we can say Firstleap, right before our acquisitions, is loss-making position. And in the past year, it is almost breakeven. And looking forward, we are still quite confident. I think maybe in the 2 to 3 years’ time, the range -- the margin could be around 10% to 20% for Firstleap. And Shunshun last year is a bit loss-making position because of the unique -- their overseas consultancy revenue recognition policy. And coming to this year, we can see some positive surprise from there. But again, because they are still very new and very young company, they need some more time to recover profitability status. But we have confidence, give them more time. They will capture and their average level of the overseas consultancy positive levels in this industry. In xueersipeiyou.com -- sorry, the Peiyou small class, the margin is quite stable and xueersi.com, which is also quite similar, is quite stable. And with more and more live classes, their margin will be a little bit better. But considering they are only around 5% of my business, so that's still not that material to my total numbers. About ASP, I think on Q1, it will be similar trend to Q4, because we did price up in general starting from summer [ '04 ] . And we will start to see the ASP increase, and this trend may continue for the other 1 or 2 quarters. But again, in the long run, we don't think the price increase is the key driver for my revenue growth. And we are a company very cautious and serious about taking price up. We prefer most of my revenue is driven by enrollments not by price. So we will be very cautious about that in price perspective.

Our next question comes from the line of Alvin Jiang with Deutsche Bank.

I have 2 quick questions. First one is on the dividend policy. And do you think going forward you will be paying at regular dividends? And the second question is double-teacher [indiscernible] . Could you give us a brief introduction of how the progress of [indiscernible] ? We noticed that you're still rolling out the double-teacher system in 9 to 10 cities at a full price. Will you consider to roll out to more cities? And then, giving out some discount in terms of this price to activate the rolling out progress?

Thank you, Alvin. The dividend policy, I think that it's a right balance between -- we need to evaluate and consider whether we have some future M&A with what's due, whether we need to invest over there. And to -- and we need to consider what's the right level of return for investments for shareholders. So we balance these 2 factors and decide whether we need to pay the dividends. I think in the filter, we will also consider these 2 factors, and decide to pay dividend or not. But by the end of the day, I cannot give you more update about that. So every year, we will evaluate these 2 factors and draft for the right balance and make decision. So we will let the class know when we want to do it again. But generally speaking, when we have more cash coming in and we have wirestream cash flow and if we don't have any future M&A we want to do in the short-term, so we will consider that. And the second question about the double-teacher model, increase on details about this call or not, something like that. I think, again, the double-teacher model is good. We are the first company in K-12 area to provide this model and we're inventing a lot of new features in this model. And -- but again, this model is still in the early stage. It work, but it still need more time to tweak on the details to make sure every detail is right. So we don't have any trend to hurry on, to accelerate the coverage of double-teacher classrooms in China. There are still ways we could well pace, control it to be and try something different, to our -- tweak our details to make it right. So we give the thing and this model more patience about that. Now specifically about price, based on what we see today, we don't give any discount on this. And again, I don't think discount is -- I'll make it price promotion or maybe something like that. The most important factors when we talk about the effectiveness of the double-teacher models, what we need to care more actually is the learning experience, learning management system, and how we can react and how these teachers, assistants, students can interact in the classroom, before the classroom, after the classrooms. So what we care more is the quality, not their price. So that's what we can see today. We still manage the double-teacher centers in very [indiscernible] , and so far, we don't have any intention to hurry up in this area.

Our next question comes from the line of Andrew Orchard from Nomura.

Couple of questions on the balance sheet. First on the long-term debt. Can you give us a sense of why you decided to raise some long-term debt, the $225 million? And also potentially related to that, can you give us an understanding of what you're doing to your short-term investments, which went up quite meaningfully in the year?

Yes, I think the debt you talk about actually is, I think, in the year -- the fiscal 2017, we have close $400 million term loan from our banks. And a $225 million outstanding term debt -- term loan, so we need to take it off right before the deadline. And the other one is the revolving loans, so we don't need to take it now right away. So the reason for us to do that is because we -- in the short-term, we don't have any huge M&A to make. But in the long run, we're still going to the global market and start off the new process, new companies, which could be very complementary to my mainstream facilities and which could be very valuable in my strategies. So we still keep as [ owned to seek that ]. And we don't want to raise the money right before we want to do a deal, because that is not right and that will cost some unnecessary fluctuations in the market. So when the conditions -- the pricing conditions are quite good, so we'll do it and will convert some of them into the time deposits. So that's also probably the reason why our short-term investments will be a little bit higher. So especially in cash perspective, we don't have anything aggressive. We don't have anything dramatic change compared to what we did in the past. Everything is right on track.

Your next question comes from the line of Sheryl Yang from CICC.

My question is regarding your revenue growth. Could you please help us to break down the revenue growth in the last quarter? Because I found out the Firstleap was something around 6% revenue in the third quarter, but only 5% this quarter. Does this mean that Firstleap is growing slower than the small class business? And what's the growth -- the revenue growth trend of Shunshun now? And what's your outlook in the fiscal year 2018?

Okay. Yes, yes. I think, firstly, for Firstleap, yes, in Q4, their growth slower than our -- it was more cash business, because that's more transactions in Q4 in small class. And in general, because -- Shunshun, that's a totally different story. Last year, they don't record that much revenue. Actually they have around [indiscernible]. And looking into to 2018, so probably the numbers will be a little bit different, will be much bigger than last year. And our -- the other mainstream business, we're looking to our one-on-one business -- actually the revenue growth for the next fiscal year is also similar to what we have seen this year. And the life will continue to be a high growth engines. And [indiscernible] our [ short fill ] small class will also be continually be very healthy and reliable. So in general, we're probably going to say still, in the long run, in 3 years’ time, we'll still be for 30% to 50%. That's the right range we can be. And the most recent year 2014 probably will -- close to the highest. So that's the key directions we can make.

And now I have a follow-up question. How should we expect the revenue contribution from Shunshun in the coming quarters?

The Shunshun revenue, actually, they start to be more meaningful when the students are [indiscernible] visa or the students are decided to go to study abroad or arrive in the country. So they have very minimal impact for my Q1. They will start to impact my revenue maybe in Q2 and Q3.

Our next question comes from the line of Lucy Yu from Bank of America Merrill Lynch.

I have 2 questions. One is that you mentioned that one more section of the spring semester class is going the first quarter of last year. So what's the impact -- actual impact of revenue and operating profit? And second one is that, I noticed that enrollment growth in fourth quarter of last year actually moderated a little bit from the third quarter. What's the reason behind that?

Okay, I think in the first place, one section, which is one weekend classes, considering in the whole quarter, we only have 30 weekends. So that's some of the [ reactions ] for you can calculate how much impact it will be. And on -- secondly, about -- your question about why enrollment is lower than the previous quarters. Frankly speaking, when we're going to the enrollment growth and revenue growth, we will see a much longer deal. What we care more is actually, the enrollment growth for the whole year, which is more meaningful. Because in the 1 or 2 quarters, some had their [indiscernible] , some -- for example, in Q4, we have one more sections, maybe by in Q2, maybe we had one less sections. So we encourage you guys to see education company in a much longer view. And specifically enrollment, we still see [indiscernible] growth across all cities, across all segments. As what I have said in my prepared remarks in the beginning, which is a little bit long, and thank you so much for your patience. But actually, I have give more clear informations about how enrollment grows across all cities. So we believe that we are still managing a healthy growth and we're trying to deliver a much reliable and sustainable growth in the long run. And also we care more about long-term than the short improvements.

Our next question comes from the line of Leon Chik from JPMorgan.

Just a few quick questions on double-teachers. Question one, are these stand-alone learning centers? Number two, are they in big cities or small cities? Number three, are they included in the 507 learning centers? And finally, I believe that only like the best teachers are teaching in the double-teacher, so why you guys worried about the quality of the education?

Thank you, Leon. You're the second one asked 4 questions. In the first place, the double-teacher actually, in Beijing, most of the double-teacher classrooms, they have 2 types. South end, they are stand-alone learning centers, for example the one in Shenzhen and [indiscernible] . But south end, actually, they are part of the current learning centers, which means for example, there are centers in [indiscernible] so they have [indiscernible] centers classrooms but they also have the double-teacher classrooms. In the long run, with this, there will be a mixed model for that. And the second things, double-teacher model for the big cities or small cities, based on what we have today, we pilot this model in the big cities, for example and Beijing and Nanjing. And in the future, we try to pilot this model in the maybe tier 2 cities. So -- which is also very important kind of pilot for us to decide how much this model can go. So while we currently we can cover both tier 1, tier 2, we don't have any plans to go to [indiscernible] tier 4 and tier 5 cities. And in the third place, your question about whether the double-teacher centers are counting in the 507 centers? Yes. The fourth question about the bad teachers teaching in the double-teacher model, why we care about their teaching quality? That's a good question. Actually, when we talk about the teaching qualities, not only the 1 hour that teacher is talking in the classroom, the teaching quality actually is decided by a lot of factors. Especially whether the students [indiscernible] per study, whether they have interactions between their master teachers, teacher assistants and the students. So they need to have more interactions in the classrooms. And it's also decided by whether we have a way to encourage them to do more homeworks or maybe to finish that -- some sections right after their classrooms. So before class, in the class and after class, we need to have established a very important learning management systems to manage the whole process. The double-teacher model is not a simple transformation from moving the teachers from physically in a position to a screen. You require a lot of change and learning management system study. We are the first company inventing this model in K-12 area. So we have a lot of lesson learned from there. So we'll continue to make it right and we are now very patient about this model. So we will let you guys know more progress when we have sent new results about the double-teacher models. Thank you, Leon.

Well like I said, practice makes perfect. So you're not going to know until you do a lot of these things. So that doesn't make sense -- much sense to be so slow.

Thank you. Ladies and gentlemen, with that we now conclude our conference for today. Thank you for participating. You may all disconnect.