Description

Internationella Engelska Skolan, aka IES, is the largest independent operator of free schools in Sweden. 20%-owned and still operated by its visionary founder Barbara Bergstrom, IES has leveraged its differentiated culture to achieve higher quality education, generating a strong brand which translates into student demand, great unit economics and large room for growth. Because of high fixed costs and burdensome regulation the business of operating schools is tough. Very tough. I think the market not pricing IES as it deserves as a result.

The summary of the thesis is…

- With very few exceptions school education is tuition-free in Sweden, meaning that the decision driver for parents and students to pick a school is based on quality rather than price

- Unlike preschool and upper secondary school, middle-school – grades 4 to 9 – is compulsory in Sweden. Incidentally 32 of the 34 IES’s schools operate in the middle school segment

- The number of students enrolling in preschool and middle school have each increased at 2% CAGR since 2011 whereas enrollment in the upper secondary school has shrunk by 3% annually

- The middle school education providers are either municipal or independent schools, with 85% and 15% respective share. Independent schools are steadily taking share from munis capitalizing on better incentives to operate sustainably

- Barriers to entry to set and run school have increased amid an increasing regulatory burden

- The independent school arena is very fragmented and prone to consolidation itself, with ~90% subscale players operating just one school

- IES’s brand recognition and efficiency have positioned the company as the largest independent operator by far 15% share amongst independents. IES will continue to steadily take share

- Despite optically not cheap IES is just selling for 10x true earnings adjusting for the fact that more than 1/5th of its schools are still immature

- In addition to the growth in Sweden IES adds the optionality of further growth in Spain

- IES benefits from a unique educational culture that yields better quality and academic results

-The company has its founder and significant owner at the helm

Brief intro to education in Sweden

Until 1992 the providers of compulsory free education have traditionally been government schools either at the municipal or at the state level (Sami schools) to a lesser extent, most of them lacking the right incentives to run operations without demanding further government support. So in order to introduce competition the government implemented a school voucher reform – a state-funded, private alternative to the public system, where entrepreneurs can set up independent schools and receive government funding based on the number of students enrolled.

The two graphs below are self-explanatory and bring up the following facts. (1) The total number of students in compulsory schools shrank in the 2000s decade and then recovered from 2010 up till now, (2) The number of students enrolled in independent schools has grown every single year regardless whether the total student population has grown or not, and (3) Independent schools have remarkably taken share since the 1992 reform, moving from less than 2% to 15% share. I was just able to get official data up to 2016, but for the purpose the graphs do the work.

Interestingly the voucher system follows an equal-treatment policy under which the reimbursement amount per student is the same no matter whether the receiver is a municipal or independent school. Truly speaking there are some complementary items for students with special needs and other subsidies for things like extra activities or homework support and reimbursement rates widely vary depending on the school location – as an example the annual reimbursement per student can vary as much as ~90,000 Kr for a school based in a large city to more than 110,000 kr for a school based in a sparsely populated municipality. But you get the main point. No voucher discrimination between the state-owned and the independent schools.

So the government essentially sets the reimbursement rates per location at the beginning of every academic year. As the regulated voucher amounts per student basically eliminate the discretionary decision to exploit pricing power, it leaves independent operators the choice to grow earnings over time by attracting more students per school, opening new schools and /or becoming more efficient. Interestingly IES operates three of the ten largest compulsory schools and nine of the ten largest amongst the independent ones, has significantly more students per school vs. peers, opens more schools than competitors and is certainly more efficient.

Given the government sets the rates per student that each school receives, how have these rates evolved over time? Leaving aside location differences amongst schools the Swedish National Agency for Education shows the average reimbursement per student in compulsory schools has compounded at ~4% CAGR since 1992, well above Sweden’s inflation rate. For IES the average voucher per student currently sits at ~95,600Kr per annum, slightly below the national average (~97,000Kr) as a direct result of IES’s schools geographical mix.

On the cost side costs per student now sit above the voucher contribution and although total costs per student have “only” grown between 3-4% per annum since the 1992 amendment as opposed to the 4% annual growth in voucher, quite a few schools still face deficits – mostly muni schools but also some independents. Unsustainable deficits in some cases, which rhymes well with the story I just exposed of tens of muni schools going out of business over the last years. On the other hand, IES’s cost base is structurally lower than those of its independent counterparts for several reasons I will later describe. This is its recipe to be profitable and steadily take share.

Even briefer description of IES history and business model  

Following the pass of the Independent School Reform in 1992 - whose main practical implication is the equal-treatment voucher-system – Barbara Bergstrom and a teaching colleague founded  IES in 1993. In 1998 Barbara and her partner separated ways with Barbara remaining as the principal of the only school IES operated back then. A US-citizen who moved to Sweden and worked as a science teacher in a muni school, Barbara opened the first IES school when she saw the shortcomings of the muni schools in terms of quality of education and student grades. The graph below shows IES’s evolution in terms o number schools under management.

The actual growth phase started in 02/03, so I plot those metrics from then, so you can see them better.

I will make two points. (1) Just compare the student CAGR 25% for IES vs. the other compulsory school options – 8% CAGR is the average of the rest of indecent operators whereas the number students in municipal schools has declined indeed. (2) More importantly this absolute growth in student numbers has not been paralleled by a proportional growth in the number of schools, which means that the number of students per school has more than doubled. While in 2003 only 330 students enrolled on average in each of IES’s 3 schools, that metric stands now above 700.

I believe scale matters at both the corporate and school level. By operating 34 schools in various locations IES can better cope with local swings of student demand. It can also exploit a small degree of operating leverage at the corporate level by spreading costs related to central functions over a larger revenue base. However given central costs are de minimis as the company runs a lean and decentralized operation – only 33 employees dedicated to HR, Finance, Payroll, IT, Legal and other corporate roles over a payroll of 2,300 – scale at the site level is more relevant as large student bases per school mean lower overheads per student.

After the screening phase – detection of suitable real estate space and favorable local demographics – the typical school takes up to two years until it opens. Upon opening a school generally takes five years to reach maturity. The first year is typically loss-making as there are pre-opening costs like the early hiring of a principal and teachers and some smaller investments in marketing. Year two a school breaks-even. Years three, four and five the school is profitable reaching full utilization at the end. The typical schedule is as follows.

I have assumed on conservative grounds that full utilization sits at 24 classrooms per school and 30 students per class. This is somehow slightly below IES average and certainly some schools operate 27 classrooms and have 32 students per class. Moreover, this larger student base and more classrooms per unit seems to be more likely moving forward. Please forgive me if I err on the conservative side!

Unit economics

The approximate current economics per student derived from H1 2018 earnings release are as follows (IES fiscal year starts in July).

Per Student

Revenue……………………………………………………………………………….. 98,000 Kr

Personnel cost………………………………………………………………………….56,500 Kr

Other opex……………………………………………………………………………....34,700 Kr

Fixed Property Costs (Rent + Leased Furniture + Equipment)……………………..20,200 Kr

Education and Meals…………………………………………………………………… 11,850 Kr

Maintenance Capex……………………………………………………………………….2,650 Kr

Total Costs (Personnel + Other opex) ……………………………………………….91,200 Kr

EBIT……………………………………………………………………………………….6,800 Kr

EBIT Margin %........................................................................................................6.9%

However I think these numbers are misleading. The voucher amount per student just rose 2.5% vs. 4% historical average and the staff costs skyrocketed by 8% vs. 3% historically. Is this the new normal? Is IES dammed to suffer from permanent margin compression? As Hillary and Tensing faced their leap-of-faith moment as they reached the now so-called Hillary step during the first successful climb to Everest, we have reached such a time to think of IES potential worthwhile investment or not. I don’t think that IES’s step is nearly as steep as the Hillary’s one though.

Municipal schools have been running deficits mostly because of the significant increase in teacher salaries and have been compensated for it. The leap of faith here is that IES will get retroactive compensation reflected in subsequent vouchers, meaning that the increases might be even above historical average. The leap of faith might become less of it if you consider the state has a legal obligation to embrace whatever the equal-treatment regulation dictates.

Let’s analyze the 8.2% and 5.9% H1 respective rise in personnel and total costs per IES student. There are two relevant points to make here. First, the numbers need two adjustments. One, the personnel costs include a 4m Kr increase in special subsidies that are 90% covered by the government. Two, there are pre-opening costs for new schools for the 2018/19 academic year that added 7m Kr for H1 to the cost base while not adding any revenue. After these adjustments staff and total costs “only” rose by 7.8 and 4.7%. While these numbers are still above the historical increase in voucher amount and certainly above the meager 3% for this year, they are not nearly as worrisome as the unadjusted number suggest.

Second and more importantly the sharp increase in wages is the direct result of the scarce supply of qualified teachers in Sweden (not of a short supply of international teachers), a fact that confers IES model a competitive advantage. Unlike muni schools, IES's focus on the English language means that 50% of its payroll consists of young teachers from outside Sweden, with lower salaries than local teachers as foreign teachers go to Sweden to pursue an international experience within a professional organization – so they can leverage it upon returning to their home countries after several years – rather than money. To put it bluntly IES has a cost advantage, so if costs at IES have notably risen you can safely imagine, despite not having official data for the year, how things played out on that front for schools mostly consisting of local teachers. As abysmal as they seem the muni school finances might trigger two actions. Either the government significantly rises its contributions for all schools or it allows the independent schools to take the place of muni schools that prove unsustainable. Either action would benefit IES. The Swedish government budget seems to be in good shape so I do not think increasing voucher contributions is a big deal. https://www.government.se/articles/2018/04/the-2018-spring-budget-in-five-minutes/

After these considerations I think IES’s normalized economics per student look more like this.

Per Student

Revenue………………………………………………………………………………..99,425 Kr

Personnel cost………………………………………………………………………...56,313 Kr

Other opex……………………………………………………………………………..34,408 Kr

Total Costs (Personnel + Other opex) …………………………………… ………. 90,721 Kr

EBIT………………………………………………………………………………………8,704 Kr

EBIT Margin %.......................................................................................................8.7%

“Steady State” Valuation

Then given we have the normalized economics per student and the ramp-up schedule for the average school we can calculate an approximate non-growth valuation. Assumptions.

- 8 immature sites. I derived that from annual reports and talks with management, so essentially some schools which opened in 2013 and 2014 reached maturity well before the five expected years as they did not seem to be smaller schools. This makes me confident that a five year ramp up period is on the conservative side. I have used therefore five years for this calculation.

- 30 students per class. I concluded from conversations with management - and unrelated professionals -  that every class can accommodate up to 32 students without impacting quality.

- Revenue and cost growth have been adjusted to historical growth rates.

- 22% effective tax rate, the Swedish nominal corporate tax rate.

IES screens poorly because of the immaturity of almost a quarter of its schools. However, it sells for <10x 2022 owner earnings, an undemanding valuation especially given its growth prospects. I guess an arguably legitimate concern every time this “adjusted for maturity” calculations to derive owner earnings show up in any retail concept – from schools to supermarkets – is what makes a person think that the ramp up will take place as optimistically as he expects. Furthermore what about growth assumptions? Too rosy? The 161,000 registrations on the waiting list (+21% YoY and +5% since the new fiscal year) tell you a lot of what you need to know about the reputation of IES in Sweden - more on this reputation later -  and consequently its chances of materializing the full ramp up of its current premises and delivering on new openings. Sure, this 161,000 number requires a downward adjustment amid some degree of mismatch between local supply and demand – i.e. too many students on the waiting list in an area where there is not enough suitable real estate to absorb that prospective demand at least in the short term -  but still is high enough to provide margin of safety when filling schools.

Growth prospects

Management has guided for 10% annual organic growth. Considering that on the price side vouchers are very likely to grow even above the historical 4% CAGR, the remaining 6% should come from expansion of current sites and new openings. Two sanity checks. One, let’s assume no expansion of the current estate (no new classes for existing schools), a very unrealistic assumption. In that situation IES should open ~2 new schools to comply with its growth target. If you look back at the graph that shows the evolution of the number of sites you can calculate that the historical average since the growth phase started in 2002/03 sits above 2 new sites per year, with that growth accelerating to 3 sites per year for the last five.

Certainly site growth comes unevenly depending on factors like local demand and availability of space, but the relevant point here is that it has come, comes and will come. For the second sanity check look at the map below. As you can see IES has significant roll-out potential of new schools out of current locations where population growth is the highest. In addition there are also expansion opportunities into lower grades 1-3 and F-classes as well as up into upper secondary school, for which I will ascribe no value. The bottom-line is that 2 new openings every year is a truly realistic number, conservative if anything.

International Expansion

In 2016 IES acquired 50% of the shares in three private Spanish school providers in Valencia (East Spain) that together form Grupo Elians. The company does not give a lot of details of this JV but I know two things. One, the family that founded and still runs the schools, the Monzonis, have a long-standing tradition and well-deserved reputation in school operations. Two, our level of English here in Spain is… well…let’s say not optimal – thus please be forgiving with my written English – so a bilingual high-quality education provider with long tradition makes sense as a relatively small investment that serves as a “trial” for this market.

IES paid ~€5m (54m Kr at the time) for half of the JV. As the three schools collectively made ~€12.4m sales (~134m Kr) and IES’s P&L shows 5.4m kr EBIT from the JV, IES paid below 1x sales and 10x EBIT. It does not look like an egregious price especially given that unlike in Sweden, where operations run under long lease contracts, Grupo Elians owns the real estate. Property comprises >60% of the amount paid. Interestingly private schools in Spain can charge “unregulated” tuition fees so they have untapped pricing power.

In late 2017 IES acquired half of the shares of another school based in Asturias (North Spain), also owned and operated by the Monzonis. Total EV was €16m (160m Kr) with net debt of 2.5m€, so the purchase consideration was €13.5m (135m Kr) including €9m worth of real estate. Financing terms are complex but in practice IES financed its 50% via a ~68m Kr intercompany loan. This school generates ~€1m EBIT (10m Kr) so 5m Kr accrue to IES’s P&L. In addition to the four schools in Spain IES also operates a small school in the UK that is barely profitable.

Rather than making a case on the international contribution to the bottom-line – 11m kr EBIT for international operations vs. more than 200m for Sweden is mostly a drop in the ocean – international operations add three interesting points.

- Management is unlikely dumb on capital allocation and it seems not to overpay.

- Related to the first point the company is willing to run experiments in new geographies with large addressable markets but is very conscious of doing them in hand with a reputed partner and not risking significant capital rather than follow a bet-the-house approach

- IES has now optionality to benefit from cash infusions if decides to proceed with sale and lease backs of its Spanish real estate assets. Interestingly the last years in Spain have seen the flourishing of the REIT market, many of their players willing to pay rich valuations for properties. I believe IES would conservatively get at least book value for its proportionally owned real estate, that is 75-80m Kr.

Valuation

Incorporating growth into the equation – essentially an average of two new annual openings, indeed 2 new schools will open on next August - I reach to the following projections.

To transform the projections into a valuation the underlying assumptions are:

- FCF roughly matches net income (even net income is likely to be higher), which I deem sensible as (1) maintenance capex is de minimis and in line with depreciation, (2) IES balance sheet shows net cash and consequently has financial gains rather than net interest expense, and (3) IES operates under negative WC with voucher collections occurring at the end of the month before they accrue to salaries and other payables.

- Discount rate is 10% and exit multiple for the terminal value sits at 17x, pretty much in line with current market valuation. I think it is reasonable as some more schools added essentially mean almost the same growth potential as IES share is tiny and the market is very large, with better chances to successfully materialize that growth as the company gets larger and larger.

Discounting back to the present the earnings stream and the terminal value I get to 4,110m Kr EV, which needs some adjustments.

EV 4,110m Kr

+ Net cash 175m Kr

- DTLs 64m Kr

- Upfront investments for the eight new schools (400,000 Kr per school approx.) 3.2m Kr

- Loan for Investment in Spanish JV 68m kr

+ Value of the Spanish JV (I conservatively assume 10x EBIT) 10.4m Kr x 10 = 104m Kr

Therefore I get to 4,254m Kr intrinsic value or 106 Kr per share, which is 55% upside if compared to the current price using conservative assumptions.

Management and Culture

In 2016 Barbara stepped down as executive chair but continues to be fully involved in the company strategy as the vicechair. These two links do a pretty good job in showing what Barbara has meant to IES and what kind of culture she has established.

https://www.thelocal.se/20150527/swedens-tough-love-enretreneur-i-still-clean-toilets-barbara-bergstrom-connectsweden-tlccu

https://www.theguardian.com/innovation-education/video/barbara-bergstrom

Incidentally she has just published her book “Tough Love” – I have not read it as it is still written only in Swedish, but I presume the English version will be released soon. The book is named after one of the three main beliefs that permeate IES, namely command of the English Language, Tough Love – which is almost self-explanatory and relates to a safe and orderly school environment where teachers can teach and students learn – and high academic expectations and aspirations. One quantitative measure of the success of such pillars is that IES’s students fare much better than other independent schools and munis when it comes to grade achievement on nationwide tests.

All this culture has gained IES a reputation that results in the long waiting list that I described. This long waiting lists translate into larger schools, and scale allows for a better value proposition then resulting in longer waiting lines… the process is self-reinforcing.

A new CEO, Annete Rampe, joined the business in March. I do not think it will mean any kind of shift in strategy especially as Barbara continues at the helm. Some points about the CEO.

- She comes from Brunswick, where she was the managing partner for its European operations, thus she has experience in dealing with issues related to large and complex organizations.

- She owns ~100,000 IES shares (circ. 7m Kr value at current prices), which is almost 3x the salary of the previous CEO (Annete’s contractual terms will be disclosed on the next annual report but my best guess is that they do not deviate that much from her predecessor’s).

- One of the things I am not a huge fan of when it comes to executive pay structure is that CEO compensation is 100% fixed – base salary + pension benefits – and lacks a performance-based variable portion. I can deal with that especially given Barbara’s ownership and commitment.

Ownership

Barbara still owns almost 20% of the company and TA Associates, a large PE with long history in the US, is the largest shareholder now with almost 30% of shares – Barbara sold 75% of the school to TA back in 2012 to monetize part of her lifetime investment and to partner with an institution that offered lots of capital and experience in growing organizations.

- On the minus side, I confess to have disdain for most PEs as investment partners. It would not be the first time nor the last that cost-cutting initiatives and the wrong set of incentives tarnish a well-regarded culture. Moreover, there was a controversial episode concerning a TA-owned college that I find disgusting to say the least. https://www.standunited.org/petition/reinstate-brian-carroll-as-campus-president Is it the case that some TA partners lack soul? Hopefully this is an isolated event.

- As a mitigant Barbara has maintained strong links with TA historically and I think TA truly respects her and her judgment, leaving her to manage IES almost freely. On top I think the fact that she still owns one-fifth of the company means that she certainly has a strong voice.

Risks

- Growth in operating costs overtime is not fully offset by an increase in vouchers amounts

- Capital allocation mistakes in choosing the wrong locations and then being locked in long leases (15 years)

- Populist initiatives gain traction to a level they harm independent schools in some way in a best case scenario or directly ban for-profit organizations from education in the worst

Catalyst

- New school roll-out

- TA buying the company out and paying a juicy premium for it