Good evening, and thank you for standing by for New Oriental's second fiscal quarter 2018 earnings conference call. At this time, all participants are in a listen-only mode. After management's prepared remarks, there will be a question and answer session. Today's conference is being recorded. If you have any objections, you may disconnect at this time. I would now like to turn the meeting over to your host for today's conference, Ms. Sisi Zhao. Thank you. Please go ahead.
Hello, everyone, and welcome to New Oriental's second fiscal quarter 2018 earnings conference call. Our financial results for the period were released earlier today and are available on the company's website as well as on Newswire services. Today, you will hear from Stephen Yang, Chief Financial Officer. After his prepared remarks, Stephen will be available to answer your questions. Before we continue, please note that the discussion 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 involve inherent risks and uncertainties. As such, our results may be materially different from the view expressed today. A number of potential risks and uncertainties are outlined in our public filings with the SEC. New Oriental does not undertake any obligation to update any forward-looking statements, except as required under applicable law. As a reminder, this conference is being recorded.
In addition, a webcast of this conference call will be available on New Oriental's investor relations website at investor.neworiental.org. I will now turn the call over to Mr. Stephen Yang. Please go ahead, Stephen.
Thank you, Sisi. Hello, everyone, and thank you for joining us on the call. We're pleased and encouraged by our results for the quarter. Net revenues in the second quarter increased to $467.2 million, which is 36.9% growth, ahead of our expectation. Accelerated top-line growth attests to the strength of our business and our sound strategy to acquire and effectively return customers and expand capacity. Total student enrollment in academic subjects, tutoring, and test prep courses went up by 43% year-over-year to approximately 1,877,100 in the second quarter. To further tap into the booming private education market and further strengthen our leadership in the market, we also added a net of 34 learning centers in 19 existing cities and expanded into the cities of Yinchuan, Shaoxing, and Huzhou with our dual teacher model classes implemented in six new schools and learning centers.
In addition, we acquired a kindergarten in Hong Kong, extending the geographic reach of our quality education offerings. Altogether, this boosts our total square meters of classroom area by a total of 38% year-over-year. In the second quarter, we remained focused on our well-proven optimized market strategy. We are steadily working our capacity expansion across cities where there is strong growth potential and where we could increase operating efficiency. We are making sound progress enhancing our online and offline integrated standardized teaching system in our K-12 business, which contributes positively to our results in the second quarter. We are also rolling out standardized teaching system for overseas test prep business, such as IELTS, TOEFL, and SAT programs in some of the large cities in China. Our business has sustained the strong momentum from the first quarter and yielded another strong top-line performance in the second quarter of fiscal year 2018.
This is mainly driven by substantial increase in student enrollments. Our K-12 all subjects after school tutoring business augmented in the second quarter, with the revenue up by around 47% and enrollment up by around 52% year-over-year, continuing from the very encouraging growth from the recent quarters. The growth in our K-12 business can be broken down into outstanding performance from our U-Can middle school, high school, after-school tutoring business, and POP Kids program, each of which achieved impressive growth respectively. Our efforts to acquire and retain loyal customers while expanding our capacity is enabling us to capture greater market share and solidify our leadership position. Historically, the second quarter is the slowest quarter in the fiscal year. However, we achieved improvement in the utilization of facilities compared to the previous quarter. This helped lift the pressure off the margins as we remain committed to investing in capacity expansion.
We are confident that the business is on the right track to regain grounds from the impact on the margin in the previous quarter. Even more encouragingly, in the second quarter, the year-over-year decline of gross margin narrowed to 70 basis points, comparing to 280 basis points in the previous quarter. The non-GAAP operating margin is down by 150 basis points year-over-year, showing significant signs of recovery as compared to 390 basis points in the previous quarter. We remain focused on driving both top-line and bottom-line growth through enhancements in cost efficiency and utilization of the facilities. I will now turn to pricing. Per program blended ASP, which is cash revenue divided by total student enrollments, decreased by about 3% year-over-year. A few factors lie behind the fall of the per program blended ASP. Firstly, our revenue mix shifted from the overseas test-prep business to K-12 after-school business with lower ASP.
Secondly, the high-ASP VIP business slowed down in this quarter. Starting from the third quarter last year, in an effort to streamline the registration process, we began to concentrate the registration for U-Can VIP classes in the first months of the first and third quarters, rather than spreading the registration evenly throughout the year. As a result, there was a high year-on-year increase of enrollments for U-Can VIP classes in the first quarter, but lower than normal growth in the second quarter. Aside from the streamlining the registration, we also currently expect that our VIP business growth will be slower compared to our overall revenue growth in the long run, which will continue to have a dampening effect on blended ASP. Thirdly, the shortened class length of our Beijing U-Can program also dilutes the per program blended ASP.
We launched the pilot program for U-Can classes in Beijing to adjust the length and format of each session from 3 hours in-class teaching to 2 hours in-class teaching, plus 0.5 hours after-class online learning. The purpose of the adjustments is to improve student in-class learning efficiency and enable them to enroll in more classes for more subjects. The pilot boosted the average number of classes enrolled per student. Hourly blended ASP, which is GAAP revenue divided by the total teaching hours, increased by approximately 6% year-over-year. To provide a breakdown of the hourly blended rate, please note that U-Can increased by 6%, POP Kids increased by 5%, and overseas test-prep program increased by 15% over year-over-year. Looking ahead, we're confident that the decline in margin will continue to ease through the remainder of the fiscal year.
More importantly, as the business expands, we will also benefit from the greater economy of scale as we continue to make strategic investments, which will generate long-term value to our customers and shareholders. Now, we will move on to the second quarter performance across our individual business lines. Our revenue driver, K-12 All-Subjects after-school tutoring business, achieved revenue growth of about 47% year-over-year for the second quarter, driven by enrollment growth of about 52% year-over-year. Breaking it down, the U-Can middle school, high school, all subjects after-school tutoring business reported a revenue increase of about 44.9% for the second quarter. Student enrollments grew approximately 48.6% year-over-year for the quarter. Our POP Kids program, again, delivered outstanding results with revenue up significantly by about 50.8% for the second quarter. Enrollment went up about 55.1% for the quarter.
Our overseas test-prep and consulting business together reported revenue growth of about 21.1% year-over-year for the second quarter. Finally, our VIP personalized class business reported a revenue growth of about 19.4% year-over-year for the second quarter. Next, I'll provide some updates on progress we're making with our optimized market strategy. We have been focusing on expanding our capacity by investing in the build-out of our O2O integrated education system, and this continues to produce very promising results. We'll start with the offline business. In the second quarter, we added a net of 34 learning centers in 19 existing cities and rolled out new teacher model class in 3 new schools and 3 new learning centers in the city of Yinchuan, Shaoxing, and Huzhou. In addition, we acquired 1 kindergarten in Hong Kong.
Our investments increased the total square meters of classroom area by the end of the quarter by approximately 38% year-over-year. In order to capture the growth opportunity in lower-tier cities, we continue to roll out our new teacher model schools and expand our business into remote areas in China. We started to pilot the new teacher model class in select cities in July 2016. By the end of the second quarter of this year, we have tested the new offering in over 30 existing cities and 10 new cities, and we're pleased to see the increased market penetration in the markets we tapped into. We also saw improved customer retention and scalability of this new model. With these encouraging results, we will continue to deploy the strategy in the rest of the fiscal year.
With respect to our online business, $18.7 million was invested in the second quarter to improve and maintain our O2O integrated education ecosystem. Most of the investments were recorded under G&A expenses. With high customer retention rates and acquisition of new customers, we believe the investments will yield significant return and bring sustainable and long-term benefits. I will first talk about O2O two-way interactive education system. Since the launching of our U-Can Visible Progress teaching system in September 2014, the interactive education system has been deployed in all existing cities. We launched the newly revamped POP Kids English program, Wangyou, in most of the cities by the end of second quarter of fiscal year 2018. This interactive education system has also been gradually used in more and more cities across China.
The interactive education system for overseas test-prep programs, including IELTS, TOEFL, and SAT courses, was rolled out and tested in most of the major cities by end of the Q2 fiscal year 2018. At the same time, we also standardized the product offerings in the city of Shenzhen, Xiamen, and Changsha. Now, I will walk you through our progress in Koolearn.com and other supplementary online education product. Koolearn.com generated net revenue of $28.2 million, representing a 59.6% increase year-over-year in the second quarter. The number of paid users increased about 23% year-over-year in the quarter. Kudou.cn live broadcast platform achieved about 709,600 registrations in the second quarter. Donut learning apps reported over 73.5 million downloads by the quarter end. Luoshu app reported over 7.1 million users by quarter end. These developments aim to extend New Oriental's traditional offline classroom teaching offerings to online education services.
This is an important front on which we set ourselves apart from other key players in the market. With a booming market and our advanced O2O product service, we are poised to gain more market share and strengthen our hold going forward. Now, let me walk you through the other key financial details for the second quarter. As mentioned earlier, the business delivered outstanding year-over-year increase in net revenue and growth in the second quarter. Due to our expansion of capacity, operating costs and expenses for the quarter were $480.3 million, representing a 40.8% increase year-over-year. Non-GAAP operating costs and expenses for the quarter, which excludes share-based compensation expenses, were $470.9 million, representing a 39.0% increase year-over-year. Cost of revenues increased by 39.1% year-over-year to $227.3 million, primarily due to the increase in teachers' compensation for more teaching hours and the number of schools learning centers in operation.
Selling marketing expenses increased 38.2% year-over-year to $72.1 million, primarily due to increase in brand promotion expenses and selling marketing staff's compensation. General and administrative expenses for the quarter increased by 44.2% year-over-year to $180.9 million. Non-GAAP general and administrative expenses, which excludes share-based compensation expenses, were $171.6 million, representing a 39.2% increase year-over-year, primarily due to increased head count as the company expanded its network of schools and learning centers, as well as the increase in R&D expenses and human resources expenses related to the development of our online and offline integrated education ecosystem. To further align vision and interest of the company internal and internal shareholders, the company granted a total of 1.5 million restricted share units of the company to employees and directors in October 2017, with graded vesting over three years.
Because of the granting, the cost involved drove the total share-based compensation expenses up by 330.2% year-over-year to $9.3 million for the quarter. Operating loss for the quarter was $13.1 million compared to an income of $0.2 million in the same period of prior fiscal year. Non-GAAP loss from operations for the quarter was $3.8 million compared to an income of $2.4 million in the same period in the prior fiscal year. Operating margin for the quarter was negative 2.8%, compared to positive 0.1% in the same period of prior fiscal year. Non-GAAP operating margin, which excludes share-based compensation expenses for the quarter, was negative 0.8%, compared to positive 0.7% in the same period of prior fiscal year. Net income attributable to New Oriental for the quarter was $4.3 million, representing a 58.7% decrease from the same period of prior fiscal year. Turning to the balance sheet.
As of November 30th, 2017, New Oriental had cash and cash equivalents of $818.1 million compared to $641 million as of May 31st, 2017. In addition, the company had $87.5 million in term deposits and $1,522.8 million in short-term investments as of November 30th, 2017. New Oriental's deferred revenue balance, which is cash collected from the registered students for courses and recognized proportionally as revenue as the instructions delivered at the end of the second quarter of the fiscal year 2018, was $1,137.3 million, an increase of 48.7% from $764.7 million in the same period of that prior fiscal year. Before moving on to our expectations for the third quarter, I would like to take a moment to reiterate our overarching goals and priorities and our optimized market strategy. In terms of our priorities, first, we will continue to expand our offline business.
We aim to add around 20% new learning centers and expand classroom area of some existing learning centers for K-12 business in existing cities. We also plan to enter 2-4 new cities where we identify as markets with the most business opportunities. In addition, we'll continue to roll out our new teacher model schools to about 5-10 new low-tier cities in China. Second, we'll continue to leverage our investment in our O2O integration in online education offerings. In particular, we will continue our focus on product refinement and the maintenance for the O2O system for K-12 business. Meanwhile, we'll continue to revamp and roll out our O2O standardized teaching system for our overseas test prep business. We will continue to make investments, and we currently believe that total spending in absolute dollar terms in fiscal year 2018 will increase moderately compared with the prior fiscal year.
Third, we will continue to focus on driving up utilization of our facilities and cost control to drive operational effectiveness and deliver long-term bottom-line growth. As already shown in the second and first fiscal quarter, we believe that the expected acceleration of the revenue growth and anticipated boost in facility utilization in the coming quarters will mitigate the impact on the margins over the coming quarters. We will keep you updated as we move through the fiscal year. We're confident that our expansion strategy and recent incentives will drive additional growth of revenue and market share in a way that creates long-term value for all shareholders. Looking at the near term and our expectations for the third quarter, we expect total net revenues to be in the range of $591.1 million to $604.2 million, representing year-over-year growth in the range of 35%-38%.
Lastly, I must mention that these expectations reflect New Oriental's current and preliminary view, which is subject to change. At this point, I will take your questions. Operator, please open the call for this.
Thank you. The question-and-answer session of this conference call will start in a moment. In order to be fair to all callers who wish to ask questions, we will take one question at a time from each caller. If you have more than one question, please request to join the question queue after your first question has been addressed. For those who wish to ask questions, please press star one on your telephone keypad and wait for your name to be announced. To cancel your request, please press the pound or hash key. The first question comes from the line of Fan Yu of Goldman Sachs. You may ask your question.
Hi, Stephen, this is Susu. You added capacity by 38% year-on-year this quarter, faster than the previous quarter at 31%. May I know the rationale behind this acceleration? Also a question relating to your guidance, the quarter guidance. May I know the forex you applied behind this sort of quarter guidance? Since if we exclude the impact from RMB appreciation, your third quarter guidance will imply around 29% year-on-year in the midpoint, actually lower than this quarter at around 34%. May I know the reason behind this deceleration? Supposedly, second half should accelerate versus first half. Thank you.
Okay. Thank you, Fan. In terms of the expansion plan, yeah, we add 41 learning centers, schools or learning centers in this quarter. The total square meters of classroom area by the end of this quarter increased by 38% year-over-year. I think the reason that we raised the expansion plan is because we are seeing the growing momentum in our K-12 business due to the solid market demand. That means we are more confident about our new O2O product as well, and the effective operation as well. For the whole year, we plan to add 20% new learning centers, combined with the 10% new square meters we rent for the existing learning centers. Altogether is 30%. I think this will drive the potential growth going forward, not only for the second half of this year, but also for the next year.
In terms of the guidance. The guidance, I think, we're still seeing the revenue growth acceleration going forward. Don't forget, in the last year, Q3, we had a very strong Q3. It's a little bit hard comparison, but I think the K-12 business will bring us very strong enrollment growth in the third quarter. I think we will see the acceleration of the growth of top line in Q3 and Q4 as well. Thank you.
The next question comes from the line of Abby Luo of Macquarie. You may ask your question.
Hi, Stephen. Hi, Sisi. Thank you for taking my question. One of my question is, just to follow up on the revenue growth, because we do see deferred revenue growth kind of accelerated to 49% from 42% last quarter. The enrollment growth is really strong this quarter. I think as you mentioned, part is of our recruitment plan of trying to have the winter and the spring together. Just want to understand how much percentage does that part account for? My second question is regarding the Hong Kong kindergarten that we acquire. Can management share more color, like what's the consideration? How many students does that actually have a margin drag? Have we consolidated this? Just some color on the kindergarten that we acquired and the logic behind it. Thank you.
Okay. Thank you, Abby. Let me answer your second question first. The kindergarten we acquired in Hong Kong is very small one, so it's not big deal. In terms of the first question you asked about the revenue guidance. In the Q3 guidance, I think on the school level, what I mean the short-term schools, the top line growth in the Q3 and the coming quarter will be very strong. That's why I said, we're still seeing the revenue acceleration of the growth. In the Q3, we had a very strong quarter last year of the other business. We had a little bit hard comparison for the Q3. We had a great deferred revenue balance growth in the quarter end.
I think, based on the historical data, 50% of the cash revenue we reported in the second quarter will be recorded into the GAAP revenue in Q3. Okay. Is that clear, Abby?
Maybe just to clarify, does that mean that if we just look at enrollment number, how much can we attribute it to maybe the spring quarter? Yeah, the winter and the spring.
Okay. We don't disclose like that, I think the most of the enrollments we have in hand are the winter classes students. That means most of the cash revenue, we report in Q2 into the Q3, the winter courses. Okay.
Okay, thank you.
Okay. Thank you, Abby.
Next question is from Natalie Wu of CICC. You may ask your question.
Hi. Good evening, management. Thanks for taking my question. Two simple questions here. First one is that how to view VIPKid model? Just wondering, how do management see the threats and opportunities related with that in longer term? Will New Oriental launch any initiatives regarding that model? Second one is, what's the current utilization rate for different subjects? Will there be a large gap between these? If yes, do you expect the gap to close, and how many years will the process take? Thank you.
Okay. Thank you, Natalie. What I just want to say is that I think most of their business is the online verbal English training model. We have already piloted the same pattern model since the last quarter. I think we're using a different way. First, for us, the one online teacher faces two to three students at the same time. We open the online class for our POP Kids students only. That means we're not open for the students outside New Oriental. For us, we have a lot of the POP Kids enrollment. That means we don't have the acquisition cost to acquire student enrollment for online employees. Also, we fix the teachers and the curriculum by the system and content. Yeah.
We're just pilot the program in Wuhan and Beijing only now. Okay. What's your second question?
Utilization.
Okay, utilization.
About the utilization rate.
Okay
subjects.
We just disclosed the utilization rates for overall. The utilization rate of this quarter is about 20%. It's flattish compared to year-over-year. Even though we opened 41 learning centers in this quarter. I think it becomes faster and faster for us to put the students into the classroom quicker than before. I think going forward, you will see the higher utilization rate for us. Yeah.
Okay. Just one more simple question. Can I know the average enrollment per student currently?
At the same time, one student takes two to three courses at the same time.
Okay, great. Thank you, Stephen.
Thank you.
Your next question comes from the line of Mariana Kou of CLSA. You may ask your question.
Hi. Thanks, Stephen and Sisi. Just wondering if, in say, longer term, if this kind of model works out, what would we expect in a year's time, the optimal enrollment per student ratio or the courses per student? The second question is on the dual teacher model. Just wanted to see, since now we have pretty decent coverage of some cities using the dual teacher model. Just wondering if you have any early indication in terms of the financial metrics, how that compares to traditional model, and also on the learning outcomes side.
Are we seeing better or comparable results compared to the traditional model? Thank you.
Yeah. We started a pilot program in Beijing, U-Can program, to change the class length from three hours in-class teaching to two hours plus 30 minutes online learning. As a result, after the change in Beijing school, we saw the average number of enrolled classes per student increased from 2 classes to 3 classes. I think it's good for us to take more students, and each of the students will take more classes at the same time. I think going forward, we will spread it out to other cities, but it depends on the time. Okay. Your second question is about the dual teacher model. Yeah. We stepped into the 30 learning centers in existing cities for the dual teacher model, and also we have already opened 10 new cities to open the dual teacher model. So far so good.
I think the student retention rate is better than we expected. I think the one teacher can face 10 or 20 classes is successfully piloted in the last year. In terms of the financial model, I think the margin of the dual teacher model should be higher than the traditional classes, because one teacher can face 200 or 300 students at the same time. All the other costs are the same compared to the traditional classes. It is too early to say because we just did the pilot program since summer of last year, and it needs more time to spread out to more cities and more learning centers. Okay. Thank you.
The next question comes from the line of Jin Yoon of Mizuho. You may ask your question.
Hi, good morning, guys. Couple of questions. Perhaps you could give us some color how to think about margins for the next quarter and perhaps for the full year. Should gross margin headwinds remain similar to Q2 levels? How should we think about G&A excluding O2O investments? Should it be growing at similarly elevated levels? Second question, Stephen, is perhaps you could give us kind of a breakdown on your revenue guide by different classes. That'd be super helpful. Thanks.
Yeah. I think the margin guidance, as you know, even though we opened 41 learning centers, the schools or learning centers in Q2, the operating margin is down by 150 basis points in the second quarter. Don't forget, our margin was down by 390 basis points in the Q1, it's a recovery. Going forward, we believe the margin pressure will lessen and reverse in the rest of the year, because I think we expect to see the acceleration of the revenue growth and also the higher facility utilization. In terms of the margin, I think in the rest of the year, the margin pressure will lessen and reverse. I won't change my guidance of the margin in the mid long-term. Our op margin target is to get 17%-18% in next three years.
This year, because we changed our expansion plan from 10% to 30%, but I think this is a good twist for us because the market demand is there, and the O2O product is better than before for us. I think for the long term, it's a good way for us to take more market share from the other key players in the market. Yeah. Okay. What's your second question? That's okay, Jin?
The second question is-
Yeah
Can you kind of give us a breakdown of your revenue guide by classes?
Okay. You mean the guidance. Okay. The overseas test prep, 10%-15%, and the K-12 business, over 50%, and the domestic test prep is 20%, all outside is the increase. The only drag down is the adult English. It will be down by 5%-10%. Okay. The pure online, yeah, the Koolearn.com, the revenue growth will be over 50% in Q3. Okay.
Great. Thanks, guys.
Okay, thanks.
Your next question comes from the line of Alex Li of Daiwa. Please ask your question.
Yeah, thanks, Stephen. I have two questions. First, just want to follow up on Fan Yu's question. Wondering how long does the management think or intends to maintain such high speed of capacity growth? Related, what makes you so confident on delivering profitability improvement given that we are still aggressively expanding? My second question is on POP Kids. It seems that I think we are getting more aggressive on non-English subjects for POP. I was wondering what types of students we are targeting and how big you think this specific subject will be growing to. Just want to know your thoughts. Thanks.
Only one-third of the POP Kids revenue comes from these non-English courses, math and Chinese. It's grown faster than the English courses. Going forward, I think the non-English courses, the revenue grows faster than the English courses. Yeah, that's for the POP Kids. The expansion plan, I think, yeah, if you remember in the last two years, we just opened like 10%-15% new learning centers every year. Don't forget, we just pilot the new O2O product since two and a half years ago. This year, we raised our capacity expansion from 10%-30%. I think it's a good way for us to take more market share. Next year, we haven't set up the budget, but I think we will open like the 20%-25% new learning centers in the next fiscal year. It depend on the market demand.
Anyway, we're quite confident about our product and the market demand. It's a huge market. Our market share is just below 2%. It's too early. That's why we raised our expansion plan this year. Yeah. Thank you, Alex.
Yeah. Thanks.
Thank you. For those participants who wish to ask question, you may ask one question at a time. Thank you. The next question comes from the line of Tian Hou of TH Capital. Please ask your question.
Hi, Stephen Yang, Sisi Zhao. Congratulations on a good quarter and also strong guidance. Since the company is entering into a relatively high-speed expansion phase, I wonder what the company's plan for next two quarters in the fiscal year. How many new learning centers do you plan to open in next two quarters and also for next year? What kind of opening plan do you have? That's my question.
Okay. Thanks, Tian Hou. I think in the rest of this fiscal year, coming Q3 and Q4, we plan to open 50 to 80 new learning centers in total in the next six months. Combined with the 85 new learning centers we set up in the first half of the year, the total number will be 160 to 170.
Next year, as I said in the last question, we haven't finished the budget of next year, we expect to open 20% plus new learning centers next year.
Thank you.
Based on my current estimation, I think the new learning centers we set up for next year will be below 30%, because we opened a lot this year. Okay.
Yes, sure. You did open a lot.
Okay.
Thank you.
Okay, thanks.
Your next question comes from the line of Lucy Yu of Bank of America. Please ask your question.
Hi, Stephen. I've got one question on the revenue growth. In this quarter, your revenue grow at around 34% in RMB terms. Can you please break that down into firstly, the retained student from summer promotions and other students, also break down that by new learning center that has been opened within the past four quarters, and the contribution from the mature learning center that has been opened over one year. Thank you.
Yeah. Actually, we had 554,000 summer enrollments in the summer. The retention rates in autumn courses was close to 50%. It's the enrollment in the autumn class we got from the summer promotion. We don't have the details of how many students from the new learning centers and how many students from the existing ones. It's really hard for us to divide the students by two parts. What I can say is, yeah, we opened a lot of learning centers. It drives the enrollments growth up, and it spends less time for us to fill the students into the new learning centers. Typically, one new learning center, it typically takes five to eight months to get a [break-even effect] . We think about how fast we can fill the students into the new learning centers. Okay, thank you.
Your next question is from Thomas Chung of Credit Suisse. Please ask your question.
Hi. Thanks, Stephen and Sisi. I have a quick question about the revenue growth and contribution in Beijing, Shanghai, Guangzhou, and Shenzhen. A quick follow-up is about the utilization. Can I have the breakdown between the old and new learning centers in the past one year? Thanks.
Okay. I think the top five cities, Beijing, Shanghai, Xi'an, Wuhan, Hangzhou, the top five K-12 business revenue contribution cities contribute 46% of total revenue. In last 12 months, the growth rate was 37% in last 12 months. Okay. This is the revenue contribution from the top five cities and the growth rate. Is it clear?
Next question.
Go ahead.
Yes, the next question is from Sheng Zhang of Morgan Stanley. Please ask your question.
Hi, Stephen, Sisi. Thank you for taking my question. My first question is about our deferred revenue. It has a very strong growth. Stephen also said that the VIP enrollment policy has changed to first quarter and third quarter. May I take that to think that the deferred revenue has even stronger growth if we look at K-12? Maybe can you give more color on the breakdown of your deferred revenue growth this quarter? Second question is about our pilot program in Beijing. You said that this is still not the time to expand to other cities. May I understand what your key concern about to expand the program to other cities, and maybe when you think it's more an appropriate time to expand this program to other cities and to POP Kids? Thank you.
Okay, within the deferred revenue balance. Yeah, I think the actual numbers, if we don't change the VIP registration window last year, I think the deferred revenue balance will be better than the 48% year-over-year growth. I think, you'll see the deferred revenue strong balance, the strong growth of the balance. I think, it means we will have a very strong top-line growth in the coming quarter, Q3. Yeah, within the deferred revenue balance, most of the cash revenue we collect from the customers are the K-12 business. Yeah. Your second question is about the pilot program that we changed the length of the class session in Beijing school.
Yes.
We just pilot the program in the summer. I think that it's good than expected, though we need more time to summarize the advantage of this program. I think we will do more and more in more cities. Yeah. Thank you.
Okay. Thank you.
Okay. Thank you.
Next question is from Tallan Zhao of Deutsche Bank. Please ask your question.
Hi, Stephen. Most of the question have been covered by previous analysts. I have two follow-up question. You mentioned about utilization rate remains pretty much same under such a high expansion on the capacity. Do I understand correctly that for mature learning centers, the utilization rate actually increased in this quarter? second quarter, it's also a follow-up question on deferred revenue.
Okay
Break it down by how much will be collected from spring season or winter season? Thanks.
Okay. I think the utilization rates for the mature learning centers, the utilization rates of facilities still going up in this quarter. I think we have a lot of room for the mature learning centers to get improvement of the utilization rates. Yeah. I think the reason that we can't get higher utilization rates in this quarter is because we opened 85 learning centers, new learning centers in last six months. I think we will fill the students into the new learning centers quicker than before. Okay. Within the deferred revenue balance, it's really hard for us to define the students in the winter class and the spring class.
For your reference, actually, the deferred revenue by end of Q2, about 47% of it will be recognized as GAAP revenue in Q3. That's roughly how we talk about the deferred revenue guidance, and deferred revenue number, and the indicator for next quarter.
Yeah. Got you.
Okay.
Thank you. Thank you so much.
Next question comes from the line of Wayne Wang of HSBC. You may ask your question.
All right. Thank you, Sisi, and thank you, Stephen, for taking my question. I have a question regarding to our capacity expansion plan in both high-tier city and low-tier city. I just want to confirm that, it seems management has mentioned that in fiscal year 2018, our total spending will remain largely flat year-on-year. What could be our full-year margin guidance? What's our full-year plan for the O2O investments? Thank you very much.
Okay. As the capacity expansion, we are opening the learning centers in the 20 to 25 cities with a higher performance in last year. This is our plan. As I said, we'll open 20% new learning centers combined with the 10% new square meters we rent for the current learning centers, altogether 30%. The online investments, we spent $18.7 million on the online investment this quarter. It's a little bit higher than we expected. Actually, we're budgeted $14 million. That means we spent $5 million more than we expected. I think it's worthy because we started to spend the online investment since three years ago. We spent $57 million in the last year and $39 million in the year before last year.
I think the high investment of the online drive the highest student retention rate, and also we require the new student enrollment, and improve the teaching quality, and feedback from the students and parents are much better than several years ago. That's why I said it's worthy. We plan to spend $ 60 million-$70 million for the whole year on the online investment. It's a little bit higher than we expected. Last year was $57 million. This year, $ 60 million-$70 million. I think it's worthy. Okay. What's your question? I think it's margin.
The second question is regarding the full-year margin guidance. It seems previously we have talked that our total expanding in fiscal year 2018 could be largely flat year-on-year. I'd like to add more color on that.
Yeah. I think going forward, in the rest of the year, we believe the margin pressure will lessen and reverse in Q3 and Q4. I won't give the specific guidance for the margins in the rest of the year. Yeah.
Okay. Thank you.
Okay.
Next question is from Eric Qi of CCBI. Please ask your question.
Good evening, Stephen and Sisi. Thank you for taking my question. I have two questions. One is, you mentioned that top four cities account for 46% of revenue. I was just wondering how many percentage of learning centers are in the top five cities, and also, how many cities have you been entered into, and what do you think the potential number of the cities you may enter into eventually? The second is, what's the utilization rate this quarter, and what's the trend for the utilization rate? Thank you.
Okay. Yeah. I mentioned the top five cities contribute 46% of total revenue. In the top five cities, I think the learning center number altogether The Beijing is 105, Shanghai is 56, and Guangzhou is 32, and Shenzhen is 25. Okay. Yeah.
Yeah.
This is the learning center-
I can send you the detail after.
Yeah
the earnings call. Yeah, it's on our presentation.
Yes.
Okay. Thank you.
Okay. We are in 70 cities. I think our ultimate goal is to step into more than 100 cities. Most of the new cities we will step into by the new teacher model. Okay. Don't forget, even in the existing cities, like in Beijing, we have 100 learning centers in Beijing, but I think the maximum learning centers in Beijing will be over 150. There's a lot of room to open more learning centers in the existing cities. Yeah.
We will take the final question from Marilyn Mo of Indus Capital. Please ask your question.
Hi, Stephen. Good evening. I have a follow-up question on margin. If I understand correctly, for this quarter, our utilization rate is kind of flattish year-over-year. The GAAP revenue per teacher hour is also up 6% year-over-year. What drives the gross margin dilution year-over-year for this quarter? Also for operating margin, besides the investment in online, what other factors drive the margin dilution? Both talk about year-over-year, not quarter-over-quarter.
I think within the G&A selling expenses, I think besides the $18 million of the online investments, I think the head count was increased because we opened more learning centers and schools in the last two to three quarters. Within the cost side, because the rental is over 40% year-over-year increase this quarter, don't forget we opened 120 new learning centers and more sq m in the last three quarters. At the quarter end, the sq m are 38% higher compared to last year. That's why we got the 44% or 45% rental increase in this quarter.
The rental increase is a 44%-45% increase?
Yeah. Yes.
If our utilization rate is kind of flattish, that should be largely offset.
Yeah. Partly it's offset, but the new learning centers, we need more time to fill the students into the classrooms. It drags the margin. On the GP level, the gross margin level is just down by 75 basis points down. It's not a big number. Okay?
Okay. Still it's because of expansion. Okay, thanks.
Yes. Okay.
We are now approaching the end of the conference call. I will now turn the call over to New Oriental CFO, Stephen Yang, for his closing remarks.
Again, thank you for joining us today. If you have any further questions, please do not hesitate to contact me or any of our investor relations representatives. Thank you.
Thank you. Ladies and gentlemen, that does conclude our conference for today. Thank you for participating, and you may all disconnect.