William P. O”Neill
| 555 Eleventh Street, N.W., Suite 1000 Washington, D.C. 20004-1304 Tel: +1.202.637.2200 Fax: +1.202.637.2201 www.lw.com
FIRM / AFFILIATE OFFICES | |
| Abu Dhabi | Milan |
| Barcelona | Moscow |
| Beijing | Munich |
| Boston | New Jersey |
| Brussels | New York |
April 30, 2014 | Chicago | Orange County |
| Doha | Paris |
| Dubai | Riyadh |
| Düsseldorf | Rome |
| Frankfurt | San Diego |
| Hamburg | San Francisco |
| Hong Kong | Shanghai |
| Houston | Silicon Valley |
| London | Singapore |
| Los Angeles | Tokyo |
| Madrid | Washington, D.C. |
VIA EDGAR AND HAND DELIVERY
Larry Spirgel
Assistant Director
Division of Corporation Finance
Securities and Exchange Commission
100 F Street, N. E.
Washington, D.C. 20549
Re: K12 Inc.
Form 10-K for the Year Ended June 30, 2013
Filed August 29, 2013
Form 10-Q for the Quarter Ended December 31, 2013
Filed February 4, 2014
File No. 001-33883
Dear Mr. Spirgel:
On behalf of our client, K12 Inc. (the “Company,” “we,” “us” or “our”), this letter sets forth the Company’s responses to the comments received from the staff of the Division of Corporation Finance (the “Staff”) of the Securities and Exchange Commission (the “Commission”) by letter dated April 10, 2014, with respect to the above-referenced annual report on Form 10-K (the “Form 10-K”) and quarterly report on Form 10-Q (the “Form 10-Q”). The Staff’s comments are set forth below in bold, followed by the Company’s response to each comment. Previously, the Company received comments from the Staff by letter dated March 4, 2014, and the Company responded by letter dated March 28, 2014. In connection with the Company’s response to each comment the Company acknowledges that:
· the Company is responsible for the adequacy and accuracy of the disclosures in its filings with the Commission;
· the Staff’s comments or changes to disclosure in response to the Staff’s comments do not foreclose the Commission from taking any action with respect to the Company’s filings; and
· the Company may not assert the Staff’s comments as a defense in any proceedings initiated by the Commission or any person under the federal securities laws of the United States.
Comment
Form 10-K, Item 1. Business, page 5
Academic Performance, page 14
1. We have considered your response to comment 1. Please clarify what disclosure you intend to provide in future filings regarding student persistence rates from year to year. In addition, please address in your Management Discussion and Analysis the trends in your student persistence rates, how persistence rates impact the number of new enrollments you need to maintain and grow your average student enrollment, and the costs related to marketing and advertising used for the recruitment of new students.
Response
K12 intends to disclose in future filings student persistence rates across all its managed public schools based on a snapshot of current enrollment, as of the most recent practical date, of students enrolled less than one year, between one and two years, between two and three years, and for three or more years, as provided in our March 28, 2014 response to comment 1.
In our Management Discussion and Analysis in future filings we will explain that we focus on persistence primarily to achieve greater academic progress for students using the K12 curriculum, as much as we do to maintain enrollment and for future overall growth. The longer a student persists with our K12 curriculum at one of our managed schools, the greater the likelihood of a better academic outcome for that student, in our experience. See K12 2014 Academic Report at pp. 4, 17.
Please note that persistence rates for schools of choice as opposed to persistence rates for traditional, local public schools will generally not be equivalent because families are free to enroll, re-enroll or withdraw their children in a virtual school as their individual needs evolve. Families are not constrained with schools of choice by public attendance zones as is often the case with public brick and mortar schools. If a child is not enrolled in an authorized alternative school like one of our managed schools, the child must attend her traditional local school. The Company continues to believe as the choice of a virtual school model becomes more widely available to families due to favorable legislative and regulatory developments, and the value of the virtual school alternative for many students is better understood, that new enrollments each year will sustain K12’s growth. We note further that, of the approximately 55 million students in grades K through 12 in the U.S. in any given year, less than 0.3% are served by our K12 curriculum. Thus, while persistence is a factor, especially as it relates to academic performance, it cannot be viewed in isolation when assessing all the other factors described above that drive our ability to increase student enrollment year over year at the managed schools we serve.
The Company does not budget marketing and advertising expense based on anticipated persistence rates for the next school year. Rather, we budget marketing spend based upon past experience and perceived efficacy of various types of media and marketing campaigns, from feedback provided by the school boards that contract for our services as well as from our enrollment centers, and management’s assessment of growth opportunities at schools that only recently have contracted for our services. Please note that, to the extent higher persistence rates
could mean that our marketing costs could decline over a prior year, it also means that we could have more available to spend toward increasing public awareness in existing and new states of the online public school option offered by the charters and school districts that contract for our services. Because overall growth in enrollment, including growth at particular schools, must be accompanied by up-front investment by us in the recruiting and training of certified teachers and the procurement of student school supplies (computers, text books and teaching materials), we carefully plan the amount of marketing to be spent during any given enrollment season against forecasted growth in our overall annual budget for each school year.
Comment
Form 10-K, Item 1. Business, page 5
Academic Performance, page 14
2. We have considered your response to comment 2; however, we continue to believe that you should address our comment in your disclosure. Please revise your disclosure in future filings to provide additional information about the academic performance of the schools you operate. For example, for each state in which K12 operates a school, please disclose a comparison of the K12 student scores in reading and math to the state averages.
Response
Initially, with respect the original comment 2 seeking comparison of K12 student scores in reading and math to state averages, we note the Staff’s statement that in connection with expanding student enrollments, it “would appear that your potential and current customers would take into account academic performance in deciding to use your products and services….” It is important to recognize that parents and families seeking to enroll their students in K12-managed schools are not our customers, as the schools we serve are public schools and tuition free. Rather, the customers we invoice are the not-for-profit school boards which have a charter from a state authority, and in some cases our customer may be the school board or superintendent of a state-authorized school district.
In selecting K12 to provide online curriculum, technology and management services, the governing charter school boards/districts perform due diligence on K12 and exercise their authority to enter into contracts with terms and conditions to meet the needs of the school and the students they seek to enroll. We compete with other providers of online curriculum and of management services to win the business or renewal contracts with these customers. As none of our direct competitors who seek to provide curriculum and management services to virtual public schools are publicly-traded companies and do not publish academic performance data in SEC periodic reports, these potential and existing customers look to the vendor applications or state education websites for academic performance data in their selection process.
More fundamentally, the publicly available information through state department of education websites and our annual Academic Reports provide investors and customers with a comprehensive, highly detailed picture of the academic performance of students attending our
managed schools. Adding substantial, granular data on test results to our SEC filings on a periodic basis will not alter or add to the “‘total mix’ of information made available,” TSC Industries, Inc. v. Northway, 426 U.S. 438, 449 (1976). Indeed, including test data from each school and each student cohort or grade would require providing detailed information on student demographics and the academic starting point of the student population at each school for a school-by-school comparison to have any meaning.
State testing information in reading and math (and in some cases, science and the English language arts) along with student demographics is publicly available for all schools, including K12 managed schools, traditional public brick and mortar schools and other charter schools, through state department of education websites. It is important to note that there are no arithmetic means or “averages” as reported by state testing authorities — rather, the state tests are intended to measure a percentage of proficiency in reading or math of the students attending a school. For example, one school may have a proficiency score of 75% in reading, and another school a score of 55%, meaning that according to the test results at a given point in time, 75% of the students in one grade or cohort have achieved reading proficiency based on a state objective at the first school, whereas 55% have achieved reading proficiency at the other school. See, for example, the Testing and Accountability web page for the State of Texas Education Agency at www.tea.state.tx.us, and click through to “district” and “campus” to see state proficiency results at every Texas school. For California, see www.cde.ca.gov and click through to the Testing and Accountability web page to find the proficiency test results for all California schools, including the K12 California Virtual Academy. For a further example, see www.isbe.state.il.us for proficiency report cards on all Illinois schools, including our Chicago Virtual Academy.(1) Families, school boards and investors can readily find this information, and importantly, look at proficiency results and student demographics from prior years for any school and make their own comparisons.
In addition, because we publish an annual academic report which now contains school proficiency data under each state’s accountability standards (which have been substituted for AYP measures in 42 states and District of Columbia who have been granted NCLB waivers by the U.S. Department of Education), we believe our customers and investors have at their immediate disposal more extensive information about the academic performance in math and reading of the virtual public schools we serve than that of schools managed by our competitors. We have also added an in-depth section to our 2014 Academic Report that provides value-added (growth) results for students in two of our larger and mature schools (Ohio Virtual Academy and the Agora schools in Pennsylvania). We respectfully refer the Staff to the K12 2014 Academic Report, now published and available at www.K12.com, as well as to our previous response to comment 2.
(1) We are supplementally providing the Staff a hard copy of the 2012-2013 “Report Card” on our Chicago Virtual Academy downloaded from the Illinois Department of Education web site at http://webprod.isbe.net/ereportcard/publicsite/getReport.aspx?=2013dcode=150162990225C-epdf.
During the 2012-2013 school year we managed some 66 schools in 32 states and the District of Columbia. State proficiency results are available on each of these schools through state department of education websites as explained above. Moreover, we have added, in our view, a more meaningful assessment through publication in our annual Academic Report of Scantron reading and math results that measure academic gains (not just proficiency) from Fall to Spring for students attending our managed schools. To add extensive data to our periodic reports filed with the SEC will not add to the “total mix” of information on our business or financial results, and indeed could lead to investor confusion.
Comment
Form 10-K, Item 1. Business, page 5
Academic Performance, page 14
3. We have considered your response to comment 3. Please provide the substance of your response in future filings.
Response
K12 will provide the substance of its response to comment 3 in its future filings. We also direct the Staff to pages 30-31 of our 2014 Academic Report which sets forth the components of each state’s school accountability systems under the waivers granted by the U.S. Department of Education under the No Child Left Behind Act.
Comment
Form 10-K, Item 1. Business, page 5
Academic Performance, page 14
4. We note your response to comment 4. Please clarify whether you intend to provide the substance of your response, including information responsive to comment 4, in future filings.
Response
K12 will provide the substance of its response to comment 4 in its future filings.
Comment
Form 10-K, Item 7. Critical Accounting Policies and Estimates, page 63
Revenue Recognition, page 64
5. We note your response to comment 6 and your proposed disclosure stating that the pro rata amount of revenues to recognize in a fiscal quarter is based on estimated total funds that each school will receive in a given year. Please further clarify how you considered
typical variations in virtual school funding formulas and methodologies in your measurement and timing of revenues (including but not limited to the following):
· Funding contingent upon successful course completion. For instance, in Florida and Texas, full-time virtual schools are allocated funds based on the number of students completing courses but receive funds only after students have successfully completed a course. Refer to ASC 605-25-30-5.
· Funding adjusted for factors such as average daily attendance and student needs.
· FTE funding based on student location.
· Funding based on prior year enrollments.
Response
Each state or school district has variations in the school funding formulas and methodologies, which we use to estimate funding at our respective schools. As we build the funding estimates for each school, we are mindful of the state definition for count dates on which reported enrollment numbers will be used for per pupil funding. Enrollment counts and definitions vary by district and may be based on a single count date, multiple count dates within the school year, and average daily attendance or membership.
The estimates we make each month on a school by school basis consider the latest information available to us and consider material relevant information at the time of the estimate. States may update their funding levels or methodologies mid-year; therefore, to the extent there are changes from historical experience we will update our estimates.
Specifically:
· Certain of our schools incorporate student progress measures that may reduce funding at a school. Texas and Florida are the only schools where funding is 100% contingent on course completion. For these schools, we look at historical completion or progress rates and current period trajectory of progress in assessing the likelihood of completion to determine estimated funding. Based on our historical experience and continuous in year monitoring, we believe we have the ability to make reliable estimates on progress measures that may impact our funding. We serve a number of schools where funding is based on some average daily attendance measure. For those schools we look at historical and current trends for student attendance and estimate the impact on the year. For example, to the extent a school has a recent trend of higher withdrawal rates than historical years, we would adjust the estimated funding for the year based on that recent information to the extent we believed it was the most reliable available.
· Special needs funding is based on specific formulas on a school by school basis and again we look at the specific requirements for providing funding information to determine the basis for the funding estimates we make.
· Funding that is based on student location is tracked through our systems and we apply the funding rates applicable to that location given the most recent available information.
· The calculation of current year funding based on prior year enrollments are essentially limits or caps to the enrollment funding formulas the states use. For example, to the extent a state enrollment shrinks greater than a specific percentage year over year, the reduction in funding from that state may be limited to a maximum decrease based on the prior year enrollment figure.
· In each of the cases above, our actual reported enrollments are subject to audit and/or verification by the states. As noted in our prior response, aggregate funding estimates differed from actual reimbursements by approximately 0.2%, (0.1%), and 0.7% of total reported revenue for the fiscal years ended June 30, 2013, 2012 and 2011, respectively.
Comment
Form 10-K, Item 7. Critical Accounting Policies and Estimates, page 63
Revenue Recognition, page 64
6. It is unclear why you stated in your response to comment 1 that you do not regard attrition (i.e., withdrawals, churn) throughout the school year as a significant risk to your business or results of operations if in fact, in some of your schools, funding is contingent on the students’ successful course completion. In estimating pro-rata amounts of revenue to recognize during each quarter, tell us the following:
· If your quarterly revenue estimate “with consideration to actual enrollments and expected funding” will yield the same result as one based on actual students (still) enrolled since the beginning of the year (or as of each state’s applicable measurement or count date ) and expected to complete the course.
· How you define withdrawals and if you factor an allowance for withdrawals in estimating revenues, particularly in your largest schools in Ohio, Colorado, and Pennsylvania which posted student churn rates as high as 51% per an NEPC study dated May 2013. If not, please tell us why.
· If you are able to estimate an average course completion rate based on historical attrition in states or districts where funding is contingent on students’ course completion.
Response
The following table sets forth total average enrollment data for students in Managed Public Schools, as previously reported in our quarterly and annual reports as indicated below.
|
| Fiscal Year Ending, |
| ||||
|
| 2013 |
| 2012 |
| 2011 |
|
September 30, |
| 121,929 |
| 106,665 |
| 77,114 |
|
December 31, |
| 119,132 |
| 104,836 |
| 75,280 |
|
March 31, |
| 118,717 |
| 105,828 |
| 75,816 |
|
June 30, |
| 117,563 |
| 104,289 |
| 74,755 |
|
To further clarify, overall attrition rates can impact revenue recognition in schools that have average daily enrollment funding as well as schools that have requirements for successful course completion or other enrollment funding mechanisms that take into consideration attrition during the year. In short, we have not seen a material change in the rate of net withdrawals within any particular school year as demonstrated by the table above. We estimate the appropriate amount of revenue to recognize is at the school level and, as explained above, we estimate based on our historical and current trends for each school.
In response to the first bullet:
· Each school is calculated differently and it would be inaccurate to provide a general response to the question. There are states where the funding would be different depending on whether a student is active throughout the year and completes the course vs. some states that have a single or multiple count date(s) and funding is provided for that student as long as they are a student on the applicable count date. Some states, on the other hand, do not have additional funding requirements relating to attendance or course completion.
In response to the second bullet:
· We define enrollment and/or withdrawals criteria in accordance with state governed policies which vary state by state and may be validated and/or audited by each state. For those states where the funding formula is dependent on some estimate of withdrawals, we do factor in an estimate depending on the criteria so that our estimate is as accurate as possible given available information to us.
In response to the third bullet:
· Yes as described above, we do estimate course completions based on historical attrition as well as completion rates in those states where the funding formula contemplates course completions.
As we explained previously and in our original response to the March 4, 2014 Comment Letter, the total number of students at the beginning and end of the school year historically varies from quarter to quarter by between 1%-4%. The referenced NEPC statement is simply incorrect because their “churn” calculation appears to be based on the number of students enrolled at the beginning of a school year compared to the number that withdraw after the school year has been completed, which may be due to high school graduation, completion of eighth grade at a K through Eight school, families that move, or students who elect to return to a traditional school or
an alternative online school.(2) Further, the reference to “posted churn rates” (posted by whom? based on what data?) is meaningless when it comes to K12’s revenue recognition policies. As discussed in our prior response, each quarter we assess the likelihood and amount of funding for each of our managed schools to determine the pro rata amount of revenues to recognize in a fiscal quarter. This is based upon actual withdrawals by students during a school year, and where funded, new students who may enroll in a school at mid-year or mid-quarter.
Comment
Form 10-K, Item 7. Critical Accounting Policies and Estimates, page 63
Revenue Recognition, page 64
7. We note your response to comment 10. Please expand your disclosure in the penultimate paragraph on page 64 to distinguish your revenue recognition policy in contracts “where K12 assumes full economic, funding and operational risk” from contracts where “K12 receives a management service fee, net of absorbing any school operating losses.” Additionally, disclose the nature of expenses that are weighted more heavily to the beginning of the year for the latter contract type.
Response
We will expand our disclosure in our future Form 10-K filings to distinguish our revenue recognition policies with respect to our two different types of contracts, and the nature of expenses that are weighted more heavily to the beginning of the year for the later.
Comment
Form 10-Q, Item 2. Comparison of the Six Months Ended December 31, 2013 and Six Months Ended December 31, 2012, page 24
8. With a view towards greater transparency, please discuss in an Executive Summary and in your management’s discussion and analysis, your revenue growth and earnings outlook based on identifiable trends in your schools’ student performance, average daily attendance, graduation rates, governance and other such factors that may affect, or have affected recent contract renewals (or non-renewals) and charter applications. For instance, we note that the Lawrence School District in Kansas has cancelled your contract because of a low 26.3% graduation rate.
(2) To calculate an attrition (withdrawal) rate during a school year, one must look at the total number of students who could withdraw to the actual number of withdrawals. For example, a K12 managed school may have an enrollment of 5,000 students on the October 2 count date, with 500 students who withdraw during the year and 500 students who are in-year enrollments, thus yielding a withdrawal rate of .09% (500/5,500). We also fail to understand the Staff’s reliance on an NEPC report as a basis for the Company’s disclosures, as the NEPC is merely one of many policy institutes — often affiliated with or funded by organizations with a particular viewpoint - that engage in the school choice debate, and that clearly disfavor the choice of full time virtual schools.
Response
We have not historically had contract renewals or non-renewals result in a materially adverse financial impact, and we have not experienced nor do we foresee a trend in non-renewals of our contracts that would have an impact on future revenue growth or earnings. We believe that we appropriately disclose the associated risks of non-renewals and charter applications in our “Risk Factors” in our Form 10-K. To the extent that a given period is impacted materially by a contract non-renewal or charter revocation at one of our managed schools, we will provide additional disclosure in our management discussion and analysis to that effect. In addition, to the extent we provide forward looking statements and guidance regarding revenue and earnings outlooks that are or could be materially impacted by a contract non-renewal or charter application, we will enhance our disclosures.
To clarify, the statement that the Lawrence School District cancelled its contract with K12 due to a low graduation rate is not accurate. Rather, K12 and the District have mutually agreed to renew the contract to cover grades K-8 but not the high school grades for the 2014-2015 school year because the District has yet to decide whether to continue offering a virtual high school alternative to students on a vendor-managed basis. The total contract, including services on behalf of the high school and grades K-8, represent less than 0.2% of our projected revenue for this fiscal year. The virtual high school enrollment for the 2013-2014 school year was approximately 290 students. There remain multiple virtual high school options for Kansas families, including a Connections Academy school sponsored by the Elkhart, Kansas School District.(3)
We note further that the 26.3% graduation rate reported by the Lawrence Journal World for the Lawrence virtual school is based upon the way the State counts student withdrawals — if a student transfers to another school to complete high school at any time during his or her four years and the District is unable to report where the student transferred or reports late, the student is deemed to have not graduated. We note also that the Lawrence School District (home of the University of Kansas) has a high graduation rate from its brick and mortar school, and a lower graduation rate from its virtual school alternative (available to students across the State) serves to dilute the District’s overall reported graduation rate.
Comment
Form 10-Q, Item 2. Comparison of the Six Months Ended December 31, 2013 and Six Months Ended December 31, 2012, page 24
9. Please also discuss the financial impact of shortfalls in your projected permanent enrollment arising from an emerging trend among districts that shift away from full-time student programs to more blended part-time programs. In this regard, we note Mr. Tim Murray’s statements in your second quarter 2014 earnings call.
(3) Connection Academy is a nationwide, for-profit privately held company owned by Pearson PLC, one of our direct competitors.
Response
Mr. Murray’s comments were made in context of “a customer’s point of view” to the previous question regarding our institutional business and licenses to use our curriculum. Our institutional business involves sales or annual licenses to school districts that manage their own virtual schools and retain the financial risks of their schools. Such districts may order services from us, including curriculum services and teaching tools, but because we do not count students served by such schools toward our total enrollment of students attending schools that we manage, we do not foresee trends in our institutional customer base affecting future enrollment.
* * * * *
Please feel free to contact me at (202) 637-2275 or Howard D. Polsky, General Counsel and Secretary of K12 Inc., at (703) 483-7158 if you have any questions regarding our responses above.
| Very truly yours, |
|
|
| /s/ William P. O’Neill |
|
|
| William P. O’Neill |
| of Latham & Watkins LLP |
cc: Kathleen Krebs, Special Counsel
Emily Drazan, Staff Attorney
Howard D. Polsky