Description

Kroton ticks almost all the boxes of a wish list for value investors: high barriers to entry, high quality business run by best in class management, strong secular tailwinds, low capital requirement, high return on invested capital, strong cash generation, expanding margins and all of the above at 8x P/E forward! The stock is liquid (c. $40m per day) and well covered (16 analysts, including GS, MS, CS, UBS, CITI). The one box we couldn’t tick is geographic exposure: 100% of Kroton business is in Brazil which happens to go through one of its worst recessions ever.

The interesting aspect of the story is that the actual recession per se should have virtually zero impact on Kroton underlying fundamentals. According to a study done by Parthenon, a leading advisory firm for the education sector globally, annual household expenditure on post-secondary education did not decrease, or even moved from its long term trends, in recessionary periods in developed economies. Post-secondary enrolments in the US have continued to demonstrate strong growth in economic downturns. In fact, the finding from a regression analysis in the US over the last 30 years was that post-secondary enrolments are countercyclical – changes in unemployment rate and duration of recessions explain 90% of the variation in enrolment growth. This actually makes sense: a potential student in doubt whether to start university or go to work will probably chose the former option if he can’t find any jobs available, which is what happens in recessions.

The reason Brazil is problematic for Kroton is rather its partial reliance on a government funding scheme, called FIES, which is now under pressure due to the government budgetary constraints. We will analyse the FIES program in some detail below but this is the source of opportunity: the market is panicking about what impact will the current economic crisis have on FIES and, in turns, what impact will a reduced FIES program have on Kroton. We shall elaborate on this below but if you are in a rush, we give you the answer now: the impact is relatively small and more than priced in the stock.

Business description

Following its recent merger with Anhanguera, Kroton is the leading post-secondary education provider in Brazil with an estimated 6-7% market share in the on-campus segment with c. 450,000 graduates currently enrolled. The on-campus segment represents c. ¾ of group sales and 2/3 of group EBITDA. The second biggest segment is the distance learning business which is the crown jewel division for Kroton. Given the size of the country as well as its sparse population in remote areas, distance learning has grown to become a very important part of the business. It represents little over 20% of sales but almost 30% of EBITDA given its high margin profile. Finally, Kroton has a small K-12 business which is currently not an important driver of value.

The business is run by Rodrigo Calvo Galindo, a well-respected and relatively young (44) CEO that was able to grow Kroton into the biggest university group in Brazil exhibiting the best financial metrics from a cash flow and margin perspective. Galindo executed his growth vision both organically and through M&A to perfection. On a consolidated basis, he was able to grow the top line at 26% CAGR from 2010 to 2015 and expanded EBITDA margins from 18% in 2010 to 41% in 2015. The business is highly cash generative and management will effectively return to shareholders the vast majority of fcf generated going forward via buybacks and dividends.

The business is high quality for the following reasons:

·         It has strong demand characteristics and it’s exposed to long term secular drivers (more on this below)

·         It provides a real service to its clients (the students) and creates for them real value (higher earnings stream potential) which leads to strong pricing power. Over the last 5 years Kroton exhibited pricing power above inflation

·         The business model is fairly steady and predictable. There is a graduate base that can be modelled relatively easily by adding to it its annual inflow of new students (a secularly growing number) and reducing it by graduates and churn (i.e. dropouts)

·         Capital requirements are typically low and cash generation is therefore pretty high

·         Barriers to entry are high. There are 2 main barriers to entry: size and brand

o    On the size front, it is important to have a large number of campuses and students for a number of reasons. There are many fixed costs in the business (marketing, curriculum, central costs) that can be levered off a larger base. Furthermore, once you build a campus, the incremental margin in filling a particular class at 60% from 50% occupancy is nearly 100%. The bigger your student base, the more chances to maximise efficiency in the student base

o    On the brand side of things, it’s crucial to have a good brand to attract students. Students are willing to pay for education if they think their degree will matter in life. The more established the brand, the higher perception of quality it will have with prospective students. Kroton happens to rank very highly in terms of quality. According to a government quality index, Kroton campuses score “good enough” to “very good” in 80% of the cases which compares to approximately the  same percentage in the public institutions (which in Brazil are perceived to be the absolute best but there are very few of them) and to 68% for the average private institution

·         High operating leverage – because a good proportion of costs are fixed in nature, especially in the on-campus segment, incremental top line revenues comes at very high incremental margin

The distance learning has grown even faster for Kroton. Following the acquisition of Anhanguera in 2015, Kroton has approximately 40% market share in distance learning with c. 520k students.

From a geographic perspective, Kroton is mainly exposed to the South of Brazil and the Mid-West which are the richer regions. Whilst growth in these regions is lower than in the North and North East of the country, these regions also come with better quality of students. As we shall see, a big focus of the government is the improvement in education levels and those campuses with lower than average scores may face some additional scrutiny. From this point of view, Kroton’s positions itself as a better than average education provider.

Kroton historical growth was partially organic and partially driven by a perfectly executed M&A strategy:

The Brazilian market remains extremely fragmented and we believe there will be more M&A opportunities in the future:

Structural / secular tailwinds

We expect the post-secondary education growth in enrolments to continue for many years. Over the past decade, enrolments grew at c. 5% CAGR in the on-campus segment whilst the distance learning segment showed explosive growth. We expect growth to continue for a number of reasons.

First and foremost, post-secondary education in Brazil is under penetrated. The enrolment rate (calculated a number of enrolments as percentage of college age people, i.e. 20-29 years old) is c. 20% which compares to an OECD average of 30%. There are currently c. 8m on-campus postsecondary students and the government has set a long term (2024) target to get to 11m. This implies a 4% CAGR from now, which is just slightly below the 5% average over the last decade. This growth is not relying in future GDP per capita growth. Whilst enrolment ratios typically go up when GSP per capita grows, in the case of Brazil they are currently sitting below the curve, meaning there is a bridge to gap before we factor in future GDP/capita growth.

Brazil is committed to close the gap to other OECD countries. Whilst the FIES program, as we shall see below, is associated with the binge spending / corrupted aspect of the Rousseff government, it’s worth pointing out that Brazil dramatically lags behind other OECD countries in terms of tertiary university spending as percentage of GDP. In Brazil, less than 1% goes to university compared to 1.6% on average for other OECD countries. Neighbouring country like Chile spends 2.5%, the US almost 3%.

In terms of total education expenditure as percentage of GDP, Brazil today spends c. 7%, which is well below the 10% Brazil set as long term target.

The secular tailwinds in the distance learning are even bigger. Many students don’t have the financial and in some cases physical ability to enrol in actual campuses to they turn to distance learning degrees. These are effectively online courses that come at a much lower tuition compared to regular universities. For Kroton, the average tuition for a distance learning course is c. 185 Reais per month compared to c. 730 Reais for a university degree (in USD that’s c. $2k per year for the on-campus degree). The nice thing about this business (beside the obscene margins it generates – c. 50%) is that it’s 100% funded by private payers. Any government action to curb FIES will likely result in a spill over effect to the distance learning segment where Kroton has 40% market share. Prospective students that can no longer afford regular universities will start taking online courses.

FIES program

In order to bridge the above mentioned enrolment ratio gap to other countries, Brazil introduced in 2009 a program called FIES. The FIES program was introduced to facilitate the enrolment of students that could not afford out of pocket tuition. It is structured as a government loan to students to pay for university fees. The way it actually works, at the end of each month the government pays the value of the tuition directly to the institutions via public bonds. C. 20% of the value of the bonds is used to pay back taxes to the government (and this is also why actual tax rate is minimal as most taxes are effectively netted off against these bonds). The remaining 80% of the bonds are sold back to the government on the repurchase program. The net effect to students is that they leave the university without having paid any tuition and with a debt to the government.

For the government, this is actually an effective and cheap way to stimulate education. Kroton estimates that it would cost the government c.40k BRL a year to send a student to a public university. Vice versa, lending money to students to go to private universities should cost the government a fraction of this.

Official delinquency rates (we are still a bit early to assess the true number) are very low, around 5%. Given the benefit that would yield to the economy, the benefits of the system are expected to outweigh significantly the costs. Since its introduction, FIES had a huge success. In 2009, when introduced, 32,000 students enrolled under such program. In 2014, the number of FIES loans granted in Brazil swelled to 720k. This positive trend continued until December 29th 2014 when the Ministry of Education announced some changes to the program, sending shockwaves to the sector and causing Kroton share price to almost halve over 3 months between December 2014 and March 2015. The stock hasn’t yet recovered since then.

On the very last day of the Government fiscal 2014, the MEC introduced some important changes. 2 new rules were introduced, known as PN 21 and PN 23:

·         PN 21 – The Government announced focus on quality of education. Therefore, only students with a certain level of education, tested through a national exam (ENAM score of 450) would be eligible for FIES loans. The net impact to Kroton would be small as very few students with a score below 450 actually enrols in Kroton campuses

·         PN 23 – For calendar 2015, the government will effectively repay the universities only 8 monthly instalments out of the 12 from the bond purchasing programme. This was entirely driven by budgetary constraints (i.e. lack of cash). Those 4 months would eventually be repaid by the government to the universities after the students graduated, i.e. in years 5 and 6 from enrolment. This will have a negative working capital effect on universities

Furthermore, the government realised that the programme was getting out of hand and it became too expensive for them. We estimate that in 2014 alone it cost BRL13bn to fund the FIES programme. Given Brazil precarious economic situation, FIES loans will be limited in 2015. Eventually, we estimate that c. 315k FIES loans were granted in 2015, down from c. 720k in 2014.

For Kroton specifically, we estimate that the number of FIES students enrolled dropped from c. 109k in 2014 to c. 58k. It’s interesting to note that notwithstanding the above, on campus number of students in 2015 will actually grow from 2014, underlying the resilience of the business model.

Finally, on November 26th 2015 (i.e. last week), Valor reported that universities in Brazil received a letter from the government announcing that the November bonds will not be entirely repurchased, only 60% of those will be paid and the rest will be rolled over. Whilst the actual cash impact for Kroton is minimal, the letter re-ignited the concerns that led to the massive selloff at the end of 2014: what will happen to FIES? How will Kroton be impacted by the loss of FIES?

US for profit education Vs. Brazil

Before we explain how we see FIES playing out, we think it’s interesting to look at differences between the US for-profit education sector and Brazil. We believe the opportunity today exists also because many US investors previously invested in Kroton see the FIES headlines and conclude the same disaster we witnessed with Apollo, Strayer and Bridgepoint is about to unfold in Brazil. However, there are massive differences between US for profit universities and Brazilian universities.

First and foremost: quality. With all due respect to the University of Phoenix, it is no Harvard. Average for profit universities in the US are widely recognised at being inferior to publicly funded / not for profit ones. In Brazil, whilst public university are very well regarded and perceived to be the best, private universities maintain a good reputation. In particular, Kroton fares well under government scoring systems which put Kroton campuses on par with public ones and at a much higher level than the average for-profit university.

A second and most important point related to the above is the expected return on the investment of going to university. There are many variables needed to calculate such IRR but the 2 most important ones are 1) expected wage uplift and 2) tuition costs. In the US, the average expected wage uplift is c. 1.6x. For Brazil, it’s c. 2.5x. Please see chart on page 116 of the very good “Education at a glance 2015” for 2015 for the OCED, a useful 500+ pages overview of education in OECD countries:

University of Phoenix is below average, say 1.3x, Kroton is above average, say 2.6x. I we include different tuition levels to calculate the IRR, we’d get massively different outcomes. On page 147 of the above booklet, there is an interesting table about different IRRs from going to university in different countries (the OECD average is c. 15%). Unfortunately Brazil is not included. Another interesting way of putting this is to calculate the expected payback period. Kroton shows the estimated payback time (in months) from different degrees. An average degree has a payback time of c. 3-4 years.

The equivalent payback time from a degree from the University of Phoenix is a lot longer, probably between 10 and 15 years. This means that whilst there is a real social and economic benefit in sending students to university in Brazil, it’s very questionable whether it’s worth spending government money to send students to some for profit universities in the US. It’s no coincidence that delinquency rates are much higher for US for profit universities than for Brazilian ones. The government in Brazil is therefore much less likely to completely renege on the program because it’s fundamentally sound. Recent changes to the FIES program were not dictated by a negative view towards the for-profit education sector but simply by budgetary constraints.

Finally, and most importantly for Kroton, the private education sector doesn’t rely on FIES in order to exist. Whilst the percentage of government funded students reached almost 90% for some for-profit universities in the US, this number is much lower in Brazil and for Kroton. For example, in 2015, we expect on-campus to represent less than ¾ of total revenues. The rest is not affected in any shape or form by FIES. Out of the c. 210k students we expect to enrol in 2015 in Kroton universities, less than 60k are FIES students, i.e. only 27%. Kroton will remain very profitable even if FIES were to disappear.

Our assessment of FIES risks

FIES is here to stay because it’s a popular program that serves a real and important purpose in Brazil. However, it’s possible that due to budgetary constraints, the program will be limited in the near term. However, we firmly believe the risks are more than priced in the market at current levels.

In order to illustrate the resilience of the business, let’s look at annual enrolments in the industry under FIES. 2015 turned out to be a very hard year with FIES loans approved falling almost 60% from 730k in 2014 to c. 315k in 2015. Things for Kroton didn’t turn out to be much better: FIES intakes fell almost 50% from 109k in 2014 to less than 60k in 2015. Still, the number of on-campus students at the end of 2015 will end up being slightly up or flat compared to 2014. FIES represented only 27% of on-campus intakes in 2015. Looking forward to 2016, Kroton will introduce private student financing alternatives to fill the gap created by lower FIES:

Crucially for Kroton, they will help structuring and putting in place the financing alternatives but they will not provide the balance sheet to do it:

The funding will come from a financial institution.

If we want to think about a bear case scenario for Kroton for 2016:

•  FIES will fall another 50% from 2015 level. MEC won’t achieve 320k as suggested but only 160k, basically going back to 2011 levels
• ·   Kroton will have an impact in line with the market, so -50% from 2015 level, so from c. 60k FIES loans to 30k FIES loans
• ·        Assume that out of the 30K lost, at least 15K will come back under new private financing alternative. Note that in 2015 Kroton started a small scale a private funding scheme and got c. 25k enrolments. Adding another 15k when they fully roll out the program doesn’t seem ambitious
• ·       We assume zero increase in out of pocket students which seems very bearish

With these assumptions, we get total on-campus students decline of c. 30k from 450k in 2015 to 420k at the end of 2016. Even in such scenario we would expect a positive top line growth due to a) price increase and b) strong performance in distance learning.

On the latter point, it’s worth pointing out that Kroton’s Distance Learning segment is a natural hedge to FIES. Kroton’s on-campus market share is c. 7% Vs. 40% market share in distance learning. It’s fair to assume that some students that would have otherwise gone to university with a FIES loan are now priced out due to lack of FIES and will try to get a cheaper degree via Distance Learning. Assuming 10% of “disappointed” students choose the distance learning, from above stress case scenario it would imply that 10% of 160k students, or 16k, would go to distance learning. Kroton would probably capture 6k of those. To put into context, in our bear case scenario we assumed Kroton loses 15k on-campus students.

Valuation

Before the negative FIES news came out in December 2014, Kroton used to trade on 16-18x forward P/E. Notwithstanding the fall in FIES volumes, 2015 EPS will end up pretty much were estimates were a year ago, i.e. c. BRL 1 per share. We forecast double digit EPS growth going forward driven by distance learning growth and continued margin expansion which would put the stock on 8x 2016 and 7x 2017 earning. In the meantime, Kroton generates tremendous amount of cash which is used for a) dividends, b) buybacks and c) opportunistic M&A. There is virtually zero debt on a net basis on the balance sheet. Looking out 2 years, setting an exit price of 12x our estimated 2018 EPS of BRL1.4 per share, we see c. 90% upside from current levels.

Catalyst

2016 FIES numbers may end up below the 320k guidance given by Ministry of Education but any guidance will probably reassure the market. Uncertainty is what is causing volatility now and any clarification on longer term outlook should drive significant re-rating.