Getting paid is the asset

On 10 August the Nigerian Electricity Regulatory Commission dissolved the board of Kaduna Electric and installed an administrator over NGN 456.5bn of market debt. Buried in the order is the number that matters: the distribution company could account for only 28.2% of the energy it received.

It is the second time NERC has taken over the same DisCo in 30 months.

So here is the question I have been sitting with all week. If you are putting money into African clean energy, what exactly are you underwriting?

WHY NOW

Three things landed inside the last month, and they point the same way.

The regulator published Q1 2026 sector data showing aggregate technical, commercial and collection losses of 37.44% against a target of 16.92%, and collection efficiency of 78.95%. DisCos billed NGN 756.93bn and banked NGN 597.56bn. No distribution company met its target.

The generation companies put sector debt at roughly NGN 7.66trn as at June 2026. Between January and April they invoiced an average of NGN 213.5bn a month and received NGN 90.8bn, a settlement rate of 42.52%. The federal government has been issuing bonds against the legacy arrears, and fresh arrears keep accruing on top.

And BloombergNEF, which has no dog in this fight, reported in July that over 99% of the $2.4bn invested in Nigerian clean energy in 2025 went into small-scale solar, because high capital costs and a shortage of bankable offtakers keep utility-scale deals from closing. BNEF added that Nigeria’s 27% benchmark rate makes local-currency borrowing impractical for capital-intensive projects.

Read those together and the constraint is not hidden. Nigeria can generate power. It struggles to convert delivered energy into collected cash.

WHAT THE MAINSTREAM EXPLANATION GETS RIGHT, AND WHERE IT STOPS

The standard account of African energy is an investment gap. The IEA’s 2026 investment report puts total African energy investment near $110bn, about 3% of the global total for 20% of the world’s population, and says it needs to more than double to over $200bn a year by 2030. Climate Policy Initiative puts climate finance flows to Africa at roughly $44bn a year against $277bn needed.

All true. All useful. And it stops one step short of the thing an investor actually has to decide.

Capital does not flow into a gap. It flows into a contract it can price. The gap persists because the contracts on offer mostly cannot be priced: the counterparty is a distribution company losing 37% of its energy, the currency is a naira with a 27% policy rate behind it, and the security package is usually a sentence in a term sheet rather than a filing at a registry.

WHAT THE PUBLIC CONVERSATION ACTUALLY ASKS

I spend time reading what people say about this in the open, on forums and comment threads rather than in press releases. The pattern is consistent.

The first question on any solar financing pitch is recourse, not return. What happens if the site stops paying. Whether a charge sits over the equipment. Who holds title. Who can step in.

Site owners have a different but connected list. Warranties stop working when the installer dissolves and restarts under a new name. Monitoring access disappears when a contract is sold on. Systems get undersized, usually the battery. Owners frequently cannot name the brand of kit on their own roof.

And there is a live public discussion of the exact structure this industry uses, where a funder owns the asset for a period and then transfers it. People work out the financing logic, accept it as legitimate, and then warn each other about contract complexity and what happens if the site owner exits early.

That is a well-informed audience asking precise questions. It deserves precise answers rather than impact language.

THE EVIDENCE FROM INDUSTRIES THAT SOLVED THIS

The useful material sits outside energy.

Telecom towers are the closest structural analogue in Africa, and the numbers are stark. Helios Towers reported H1 2026 with 69% of revenue in hard currency, 98% of a $5.9bn contracted revenue book from large multinational operators, an average remaining contract life of 6.5 years, and both CPI escalators and power-price escalators written into customer contracts. IHS Towers runs a similar architecture in Nigeria and discloses that power cost pass-through lags by about one quarter. Through the 2023-24 naira collapse, MTN Nigeria’s dollar revenue fell around 60%; IHS Nigeria’s fell around 26%.

US solar securitisation supplies the structural lesson. KBRA data from March 2026 shows solar leases running cumulative gross losses of 0.5% to 2% at 24 months, against 1.5% to 6% for solar loans, with rating stability of 98% against 83.6%. Owning the asset and selling the service performs better than lending against it.

Servicing continuity supplies the credibility lesson. Sunnova and Solar Mosaic both filed for Chapter 11 in June 2025 with roughly $12bn of asset-backed bonds between them. The bonds kept paying because a back-up servicer and a transition manager were named in every deal from the start.

And pay-as-you-go solar supplies the underwriting variables. GOGLA and MFR, working across 17 country businesses, find a repayment rate of 72% of financed value at twice the contract term, with West Africa at 77% and East Africa at 69%. Two levers dominate. Cohorts with deposits above 10% outperform right through the contract, and the worst outcomes sit in cohorts below 5%. The repayment rate at 90 days predicts the endpoint: top performers near 100%, laggards below 60%, and laggards stay laggards.

THE COUNTERARGUMENT, WHICH IS A GOOD ONE

Pay-as-you-go is also the strongest evidence against the optimistic case. In 2023 around half of pay-as-you-go customers were either written off or more than 30 days late. Repayment across the sector varies from below 60% to above 90%, which says collectability is a property of the operator rather than of the technology.

There is a sharper version. The enforcement tool that makes household pay-as-you-go work is remote shut-off. That tool is blunt at exactly the sites this business targets. You do not switch off a hospital theatre because an invoice is late.

A second counterargument worth taking seriously: most pay-as-you-go performance data is self-reported into an industry monitor and is not independently audited. Treat 72% as a benchmark to beat, not as a verified sector fact.

WHERE THAT LEAVES THE POSITION

Two things follow, and they are practical.

Choose offtakers whose current alternative is diesel. Diesel ran around NGN 1,730 a litre in early August. For a site already buying it, the monthly service payment replaces a bigger bill rather than adding a new one. That is a different credit conversation from selling to a customer on a subsidised tariff.

For critical facilities, design step-in in place of shut-off. Perfected security over the equipment, a named back-up operator who can take over the service contract, and terms that let the asset change hands without the site going dark.

A CHECKLIST FOR ANYONE ASSESSING A DISTRIBUTED ENERGY ASSET

Who is the offtaker, and what are they spending on energy today? Diesel spend is verifiable. A promised saving is not.

What deposit is being taken, and how does it compare with the 10% threshold in the pay-as-you-go data?

What is the payment record at day 90? Treat month three as a formal go or no-go gate on every site.

Who holds title to the equipment, and is security perfected at the National Collateral Registry rather than described in a term sheet?

Who is the named back-up operator, and can the service contract transfer without the site losing power?

What escalators sit in the contract, what share of the payment is hard-currency indexed, and what is the disclosed pass-through lag?

Is the metering and monitoring data available to the owner and to the financier, and does that survive a change of servicer?

What portfolio-level guarantee or liquidity cover could wrap this, and has the SPV been structured so it can be wrapped?

CONCLUSION

The scarce input in this market is a counterparty who pays every month and the machinery that makes sure they do. Kit is cheap and getting cheaper. Installers are learning. The payment chain is where the value sits, and it is buildable, contract by contract, by people willing to do unglamorous work for a decade.

That is the part I want Tribes Capital judged on.

Sources: NERC Order NERC/2026/086, 10 August 2026. NERC Q1 2026 sector report. THISDAY on Genco debt, 23 July 2026. BloombergNEF Sub-Saharan Africa outlook, July 2026. IEA World Energy Investment 2026. Climate Policy Initiative, Africa climate finance report. Helios Towers H1 2026. KBRA, 31 March 2026. GOGLA and MFR PAYGo KPIs. Energy for Growth Hub, November 2025.

THREE DISCUSSION QUESTIONS

If collectability is a property of the operator rather than the technology, what evidence would persuade you that a specific operator collects well? What would you want to see published?

For a hospital or a school, what is a fair substitute for remote shut-off? Where should the line sit between protecting the investor and keeping a critical facility running?

Which of the four mechanisms above, escalators, retained title, named back-up servicer, perfected security, would you insist on first, and which would you trade away?