Almost Break-Even, Still Bleeding: Safaricom Ethiopia's KSh 434M Loss Hides a KSh 16.6B BillAlmost Break-Even, Still Bleeding: Safaricom Ethiopia's KSh 434M Loss Hides a KSh 16.6B Bill

Both numbers are true, and the gap between them is the most useful thing to understand about the "break-even" target that Safaricom has promised by March 2027. The target is for a measure called EBITDA, which leaves out the costs that make building a phone network so expensive. Reaching it would be a real milestone, but it would not mean Ethiopia is making money.

Four rungs between "close to break-even" and "profit"

Think of the Ethiopian results as a ladder, with each rung adding a cost the one above it ignored. At the top sits EBITDA, which counts only the cash cost of running the business against the money coming in. That loss was KSh 434.4M ($3.34M) in the half-year results Safaricom published on 5 November.

The next rung adds the wear on the network itself. Masts, fibre and equipment lose value every year, and accountants charge that loss against profit as depreciation and amortisation. As Mwenendo reported from the results, the Ethiopian loss after that charge, known as EBIT, was KSh 11.4B ($87.75M), down from KSh 24.3B ($187.06M).

On Mwenendo's arithmetic, network wear therefore came to about KSh 11B ($84.44M) in six months.

The third rung adds net finance costs and tax, which Safaricom's results put together at KSh 4.92B ($37.87M). The last rung is the net loss of about KSh 16.6B ($127.78M). The public figures leave about KSh 0.3B ($2.31M) unexplained between those rungs, which is a reminder that the full breakdown sits in the company's accounts, not the headline.

There is a cash version of the same story. Safaricom's results show operating free cash flow of minus KSh 10.9B ($83.90M), after capital spending of KSh 5B ($38.49M), which was down 47.2% on a year earlier.

A worked example: the bus that "breaks even"

The numbers below are invented to show the logic, not drawn from any real operator.

A Nairobi operator buys a minibus for KSh 6M ($46,185) with a KSh 4M ($30,790) loan at 12.5% a year. In a year, fares bring in KSh 2.4M ($18,474) and fuel, the driver and repairs cost KSh 2.4M ($18,474). On the road, the bus breaks even. That is its EBITDA.

But the bus loses value. If it lasts six years, it loses KSh 1M ($7,697) a year in depreciation. The loan costs KSh 500,000 ($3,849) in annual interest. The owner's true result is a loss of KSh 1.5M ($11,546) a year, while the bus still "breaks even" by the measure the owner quotes to friends.

Safaricom Ethiopia is that bus at the scale of a national network. The EBITDA line is shrinking fast because revenue is growing faster than running costs. Service revenue rose 71.6% to KSh 10.6B ($81.59M), or 102.1% in birr terms, and active customers over the last 30 days reached 12.2 million. The problem is that the capital cost of the network does not shrink at the same pace.

Why the birr is the other half of the story

Ethiopia's currency has been the second source of pain. Safaricom's results show that the birr fell 4.3% against the dollar in the period, against 16.9% in the previous one. Currency losses on operating balances, which Safaricom reports separately, fell from KSh 9.39B ($72.28M) to KSh 1.07B ($8.24M).

That line is not the whole exposure. It does not cover the loans. Foreign-currency borrowing rose to $311M, or about KSh 40.4B at the Central Bank of Kenya's rate of KSh 129.91 to the dollar. When the birr weakens, a loan repaid in dollars needs more birr, while the revenue that pays it arrives in birr. A slower slide eases the pressure, but a sudden one would reverse the improvement quickly.

Safaricom has said the pace of progress depends partly on pricing and exchange rates. The figures suggest that a steady birr matters as much to the March 2027 target as extra customers do.

Three numbers that will show if it is real

The first is the EBIT loss, which should fall from KSh 11.4B as revenue grows into the network. The second is capital spending: cutting it by 47.2% helps cash now but can slow the network, which stood at 3,624 sites in September, up from 3,504 in March. The third is mobile money. Safaricom's results show that one-month active M-Pesa users in Ethiopia fell to 2.15 million in September from 2.58 million in June, the second quarterly fall in a row.

Safaricom guides that the Ethiopian operating loss for the full year will run between KSh 12B ($92.37M) and KSh 15B ($115.46M), before leftover hyperinflation accounting charges. That is the number against which the first half should be judged, because it shows the company itself does not expect profit this year.

For Kenyans who own Safaricom shares, directly or through a pension fund or SACCO, the question is when Ethiopia stops subtracting from group earnings. The Kenyan business is carrying the load: group net income rose to KSh 58.2B ($448.2M), and M-Pesa passed KSh 100B in half-year revenue, as Mwenendo's fee ladder explainer showed.

How to read any "break-even" headline

Ask three things before cheering. Break-even before what: operating costs only, or interest and network wear as well? Does the figure include currency losses, or strip them out? And is cash flow positive, since a company can report accounting progress and still need fresh loans? A company that passes only the first test is moving in the right direction, but it has not yet arrived.