Grid watch

  • 2,316 MW Kenya peak demand
  • 3,243 MW Installed capacity
  • 89 % Renewable share
  • 3.42 KSh/kWh Fuel cost charge
  • 184.5 KSh/l Super petrol, Nairobi
  • 9.7 m Households connected

Indicative figures, updated by the desk

Off-Grid

Pay-as-you-go solar, explained as a lending business

The panels are the visible part. What made solar home systems spread across East Africa was consumer credit with a remote off-switch.

Abstract illustration of solar array rows
Photo: Energy Siren

A solar home system costs more than most rural households can pay at once and less than a bank will bother to lend against. Pay-as-you-go exists to bridge that gap, and understanding it as a credit product rather than a solar product explains almost everything about how these companies behave.

The mechanism#

A customer pays a deposit and takes the system home. The unit contains a controller that requires a periodic unlock code. The customer buys credit by mobile money; the system stays on. Payments stop; the system stops. After the full amount is repaid, the unit unlocks permanently and belongs to the customer.

The remote lockout is the collateral. There is no repossession, no court process, and no collection agent walking to a village. That single design decision made lending viable at ticket sizes and locations where conventional consumer credit could not operate.

Two balance sheets, not one#

Every PAYG company runs two businesses at once.

The first buys hardware, usually priced in dollars, ships it, and holds inventory. The second issues thousands of small loans and waits to be repaid in local currency over one to three years.

This creates the structural problem in the sector. Costs are incurred immediately and in hard currency. Revenue arrives slowly and in shillings. Growth consumes cash rather than generating it, which is why rapid unit growth has repeatedly coincided with funding stress rather than profitability.

What the numbers to watch actually are#

Units sold is the metric most often quoted and the least informative. Three others matter more:

  1. Collection rate. What share of expected payments actually arrive, cohort by cohort.
  2. Portfolio at risk. How much of the outstanding book is behind schedule, and by how long.
  3. Cost to serve. Field agent, logistics and after-sales cost per active customer.

A company can ship record volumes while its collection rate deteriorates. The deterioration shows up eighteen months later, which is long enough for a growth story to outrun its own portfolio quality.

After-sales is the hidden cost#

A system sold in a remote location has to be repairable in that location. Warranty obligations run for years, spare parts must reach places with no reliable road access, and a customer whose lights failed while they are still repaying will stop paying — reasonably.

Companies that treated service as a cost centre to be minimised generally found that the savings appeared in the collections column instead.

Cheap hardware with expensive service is a worse business than solid hardware with predictable service. The repayment book notices the difference.

A sector rule of thumb

Where the model runs out#

PAYG works for lighting, phone charging, radio and small television. It works less well as customers want refrigeration, pumping or milling, because the system size and therefore the loan size grows past what the model can underwrite.

That ceiling is where PAYG hands over to mini-grids and to grid connection. Treating the three as competitors misreads them; they occupy different rungs of the same ladder.

Sources and further reading