The Most Important Asset in School Finance Is Not the Campus

Education is often described as defensive because families and governments prioritise learning. The evidence says something less comfortable: enrolment, campuses and pilots matter only when they produce dependable, collectible revenue.

Nineteen of 25 school projects were partly or fully dropped.

That is the number that should make anyone pause before calling education a defensive investment.

The projects were backed by the International Finance Corporation, the World Bank Group institution that invests in private businesses in developing markets. Between the mid-2010s and 2021, its direct K-12 school investments in Sub-Saharan Africa and East Asia totalled $156 million. Of that, $88 million in loan commitments was cancelled. Eight projects went unused altogether.

The schools were not selling a frivolous product. They were selling education, perhaps the product parents are most reluctant to abandon. Yet the World Bank's Independent Evaluation Group found the familiar culprits: enrolment below expectations, tuition arrears and unreliable fee receipts. Some borrowers also had foreign-currency loans while receiving tuition in local currency.

The IFC's school portfolio exposed how quickly strong education demand can fail to become dependable repayment. Photo: APK / Wikimedia Commons, CC BY 4.0.

The question is deceptively simple. If parents need schools and governments need educated citizens, why can a school still be a poor credit?

Because demand is only the first link in a long chain. And it is the easiest link to see.

The comforting story financiers tell themselves

The standard story goes like this: a school opens classrooms, enrolment rises, fees follow, and the physical campus provides a useful backstop. For an edtech company, the modern version is a successful pilot: pupils use the product, teachers like it, and expansion capital can fund the next market.

At first glance, this sounds sensible. A packed school or an enthusiastic pilot feels more tangible than a spreadsheet.

But a lender does not receive school demand. A lender receives scheduled payments. An equity investor does not own a classroom full of pupils. They own a claim on whatever money remains after teachers, rent, technology and debt are paid.

That distinction is not pedantic. It is the whole case.

The IFC experience is unusually stark because the institution was not dealing with a single poorly run school. It was looking across a portfolio. More resilient operators tended to charge higher fees or had diversified sponsor backing. One East Asian equity investment performed well, helped by capitalisation and international expansion capacity. The weaker cases depended too heavily on local tuition that did not arrive predictably enough.

Put those outcomes beside the cancellations and a pattern appears that none of the glossy education-investment language quite says outright: a school can be socially necessary, visibly busy and financially fragile at the same time.

The money disappears between the parent and the lender

Imagine a parent who wants their child in school. That is demand. Now imagine the parent’s income weakens, the fee is delayed, or payment is made irregularly. The child may still attend. The school still has to pay its staff. Its lender still expects instalments.

That gap is where education sector financing succeeds or fails.

Foreign-currency debt makes the gap wider. The school collects fees in local currency, but owes money in another currency. Nothing has to go wrong inside the classroom for the repayment burden to become unmanageable. A borrower can be educating children exactly as planned and still be unable to service its loan.

This is where most people stop looking. They see enrolment and assume revenue. They see revenue and assume cash. They see a campus and assume security.

Those are three different things.

The lesson also answers a question many founders ask about early edtech validation. Are five or ten users who love a product and ask for more lectures enough? They are enough to justify learning more. They are not enough to prove a financeable business.

A serious investor needs to know who has authority to pay, whether that buyer renews, whether the product improves outcomes that matter to the buyer, and whether a pilot converts into a paid recurring contract. A teacher may use a tool. A school, parent or public body may pay for it. Those are not always the same person. Confusing the user with the payer is one of the fastest ways to mistake interest for revenue.

Arizona's warning: buildings do not pay the bond

By late 2025, Edkey Charter Network, the Arizona operator behind Sequoia Pathway Academy, had fallen behind on bond payments amid declining enrolment and a financial crisis. A bond trustee advanced a $1 million loan under a forbearance arrangement, to be repaid by 28 February 2026.

The loan bought time. Edkey could consolidate campuses and pursue online schooling while state oversight loomed. But that is the revealing part: an operator with physical sites still needed emergency liquidity to bridge an operating crisis.

Edkey's emergency funding showed that campuses do not prevent an enrolment decline becoming a liquidity crisis. Photo: Abhishek Navlakha / Pexels, Pexels licence (free commercial use).

The conventional reading of school property is that it makes a business safer. Property may have value, certainly. It does not automatically cover the monthly mismatch between what comes in and what must go out.

Edkey is not evidence that every charter school is distressed. It is evidence of something narrower and more useful: falling enrolment can move quickly from a planning problem to a solvency problem when debt payments are fixed.

A lender's job is not to be impressed by the building. It is to ask whether the building produces enough dependable cash after the operating costs of teaching inside it.

That is also why an education founder's pitch deck needs more than a market-size slide and photographs of new facilities. The uncomfortable questions are better: What share of fees is collected on time? How much is overdue? Which payer accounts for the revenue? What happens if enrolment falls? Who can legally operate the service, and what permission could change?

The other Arizona story is about boring competence

A few months later, in June 2026, Paideia Academies, a nonprofit K-12 charter operator in Phoenix, took a different path. HJ Sims arranged a $20.825 million bond refinancing that consolidated three earlier bond issuances from 2015, 2019 and 2022.

The structure was expected to save more than $50,000 a year and included a declining premium call, giving Paideia flexibility if rates improved before the 2036 maturity.

Paideia's refinancing made debt structure, not new construction, the immediate source of resilience. Photo: Nyttend / Wikimedia Commons, Public domain.

There is no glamorous lesson here. That is precisely why it matters.

Paideia did not make financing risk disappear by having several campuses. It reduced strain by simplifying its debt and lowering annual costs. Finance people call this refinancing. In plain English, it is replacing a complicated set of old borrowings with one more manageable arrangement.

The contrast with Edkey is not that one American school was good and one bad. Mature capital markets can offer a troubled operator a trustee loan, and a better-positioned operator access to refinancing. Those tools give management room to act. They do not make weak collections, weak enrolment or weak controls harmless.

For any operator planning school expansion capital, debt structure deserves the same attention as the new classrooms. A cheap-looking loan that cannot be refinanced, or that demands payments before collections settle, can turn growth into a trap. As GI Network has argued in its examination of refinancing risk, the test is not whether funding is available today. It is whether the business can carry it through the next decision point.

Kenya found a better question than collateral

Ed Partners Africa, a Kenyan lender to low-cost private schools, began from a less flattering set of facts. It serves schools charging under $300 a year. Many would look informal. Many would not present the sort of property security a conventional lender prefers.

Since 2016, Ed Partners has disbursed loans to more than 420 schools serving more than 123,800 students. More than two-thirds of its loans were for classrooms. Its client entrepreneurs were 56 percent women.

Yet the important feature of the model is not the classroom loan. It is what surrounds it: financial and cash-flow analysis before money is disbursed, followed by support on fee collection and management.

Ed Partners did not assume low-cost schools were safe because education is important. Nor did it reject them as unbankable because they lacked conventional collateral. It tried to understand how money actually moved through the school.

That is a more demanding form of underwriting, the lender's process for deciding whether a borrower can repay. It requires looking at collection practices, management quality and the timing of cash, not merely the value of a building.

What surprised us was that the strongest counterexample in this research does not weaken the central argument. It strengthens it. Rising Schools, an IDP Foundation initiative operating in Ghana and Kenya, achieved repayment rates above 90 percent on guaranteed loans to low-cost private schools. It then moved towards commercial loans with technical assistance and introduced incentive-linked loans tied to learning improvements.

The message is not that every school can borrow. It is that lending can work when finance is paired with the operating discipline needed to make repayment plausible.

Pakistan offers an even sharper warning. In a randomized trial, financial or educational support eliminated school closures over four years. But loans increased closure among smaller schools, while educational support helped both smaller and larger schools. Money alone can accelerate a weak operator's problems. A lender who treats capacity support as charity rather than risk management has misunderstood the asset.

Demand survived. Nova did not.

Japan's Nova language-school chain collapsed in 2007 despite strong demand for English tuition. Delayed payments, accounting problems and fee mismanagement were central to the failure.

That case is useful because it removes the easy explanation. This was not simply a frontier-market story, or a story about low-income parents, or a shortage of school buildings. Demand existed. The payment system and financial controls did not hold.

Across Japan, Kenya, the United States, Sub-Saharan Africa and East Asia, the same truth keeps resurfacing in different clothes. Education demand may be durable. Education cash flow is conditional.

GI Network's view: The first question in education sector financing should not be, “How many learners need this?” It should be, “Can this operator turn learner need into collected cash, without betting the business on a single payer, policy decision or refinancing event?”

When the opposite is true

It would be wrong to conclude that education is inherently fragile or that physical campuses have no value. Stronger schools in the IFC portfolio did better where fee levels were higher or sponsor backing was broader. Paideia's refinancing shows that active debt management can strengthen resilience. Rising Schools shows that guarantees and technical support can create repayment performance that commercial lenders can eventually use.

The evidence does not prove that every school needs the same capital structure. It proves that finance must match the operator's actual cash pattern and management capacity.

Nor is country risk an adequate shortcut. American operators can need emergency support. Kenyan lenders can build viable models for schools that traditional finance overlooks. That is the same broader lesson behind why country risk is a poor excuse for a bad deal: labels cannot substitute for examining how a specific business gets paid.

What operators should prove before asking for money

For school operators, the work begins before the investor meeting.

First, separate enrolment from collection. Track who pays, when they pay, how much becomes overdue and whether one payer group can disrupt the whole business. A school with rising enrolment but worsening arrears is not demonstrating growth. It is demonstrating a future cash problem.

Second, match the funding currency to the income currency where possible. The IFC cases show why. Local tuition supporting foreign-currency repayment is not merely a technical mismatch; it puts the operating business at the mercy of movements it cannot control.

Third, make outcomes part of the commercial case. For an edtech provider, this means defining the result the buyer cares about before scaling the curriculum or product. A pilot should produce evidence not only that learners engage, but that the paying organisation renews and that the reason for renewal is clear. A successful unpaid pilot is a research result, not a revenue model.

Fourth, treat permission and trust as operating assets. Licensing, curriculum approval, child-data governance and safeguards around inappropriate AI outputs belong in diligence because they can determine whether a provider is allowed to keep selling. A product used by children has a higher burden of care. It should also have a higher burden of proof.

Finally, ask for capital that solves the constraint you actually have. More classroom space will not fix weak fee collection. Venture funding will not fix a business whose pilot customers have no budget authority. Debt will not fix a small operator that needs management support before it needs another repayment date.

What experienced investors see first

First-time education investors often see the mission, the campus and the enrolment chart. Experienced investors look for the break in the demand-to-cash chain.

They ask whether the buyer is the user. They ask whether payments are timely, not merely invoiced. They look for payer concentration, renewal behaviour and the exposure to delayed public reimbursement. They test whether learning outcomes support pricing power or contract renewal. They want to know whether regulation can interrupt delivery and whether data practices create risks the business has not priced.

For lenders, the psychology is straightforward. Debt has no patience for a promising story. It needs a path to repayment that works in an ordinary month, not just in the management forecast.

For impact investors, the temptation is slightly different. Because the social value is obvious, they may accept operational ambiguity that they would reject elsewhere. But the Pakistan evidence should give them pause. Poorly matched finance can damage the very schools it aims to support.

GI Network would begin by mapping the operator's demand-to-cash chain: learner need, legal payer, collection record, renewal evidence, outcome proof, regulatory permissions and capital requirements. We would then test the cash flows against the proposed debt or equity structure, identify the weak links before outreach, assemble the evidence an investment committee will challenge, and target capital providers whose mandate fits the actual risk rather than the founder's preferred label.

Use the Chain Test

Before calling an education business defensive, run the Demand-to-Cash Chain Test. Each link must hold:

1. 1.Need: Do learners genuinely need the service?
2. 2.Authority: Does the person using it have a payer with the authority and budget to buy it?
3. 3.Outcome: Is there credible proof that the service delivers the learning result that keeps the payer engaged?
4. 4.Renewal: Do pilots, enrolments or contracts turn into repeated paid use?
5. 5.Collection: Does billed revenue arrive on time and in a currency that matches the obligations?
6. 6.Permission: Can the operator continue to deliver under its licences, approvals and safeguarding duties?
7. 7.Capacity: Does the business have the management and financing structure to survive growth?

Miss one link and education demand may still be high. The school may still be full. The pilot may still be loved.

But the cash flow is not defensive yet.