Who Got the Better Deal? Judging the Safaricom Sale
The control of Safaricom changed hands with Vodafone Kenya Ltd (VKL) acquiring an additional 15% from the government of Kenya (GoK) at KES 34.00 a share in a single arranged trade.¹ GoK sold two things, a 15% stake at KES 34 a share and the future dividends on the 20% stake it kept. Money and shares have both changed hands, so the question can be answered properly, who got the better deal?
Shares Exchange
There is no argument that Safaricom is a good business to own, the discussion is whether at 34 VKL paid fair value for that acquisition. Over the last three years, SCOM's share price has dipped to as low as eleven a share, the price when the deal was struck on 4th December 2025 was 28. Three prices for the same piece of business. So, which price should VKL have aimed for? A price is only cheap or dear against a theory of the company's intrinsic worth. One way to look at this worth is by the Graham number. This measure aggregates a company's earnings, its book value, the assets and debt into one figure representing the highest price an investor should pay for a company's share. According to this theory, VKL should have paid √(22.5 × 1.53 × 4.61) ≈ 13; also P/E 19.5×, dividend yield 5.9%View full workings (A) →. However, Graham numbers theory prices a company on the one number quality makes irrelevant. Book value carries equal weight with earnings. Book value measures what was spent building the business, not what the business is worth, and Warren Buffett would agree. If Buffett was looking at this deal to affirm if the 34 is a fair price, he would question if the owner-earnings yield, plus the growth you are betting SCOM will experience, comfortably beat the long government bond. At 34, SCOM's earnings yield is about 5%. If VKL invested the same amount of money on government debt, the investment would pay 13.6% on the Kenyan 20-year bond that auctioned the week of the deal.² That alternative starts 8.5 points ahead of the SCOM investment. Buying into SCOM instead of bonds is betting on the company's annual growth to close that gap for the same returns. Likely, Buffett would weigh the SCOM investment as a wonderful company at a price that only works if the future cooperates. But that test weighs only today's earnings against a bond. On today's earnings, 34 barely wins. Whatever makes it a good price is not in what Safaricom earns now, but in what it will earn next.
The Dividends Sold Early
GoK also sold its dividend rights in exchange for an advance. The deal was the treasury keeps its 20% stake and the voting rights, but VKL gets the right to collect future dividends up to a maximum of KES 55.7 billion, for KES 40.2 billion paid now.¹ Once VKL collects up to that maximum, the dividends return to the government. This advance is essentially a loan and its rate is tied to what dividend is paid in the future. If the dividend holds at the current 2 shilling, the loan will clear in 3 and a half years at KES 40.2bn now for up to 55.7bn; ~16% if held at 2.00, ~10.5% if it falls to 1.20View full workings (C) →. If the dividend falls back to its 1.2 level of 2024 and 2025, VKL waits nearly six years and earns about 10.5%, and if the dividend improves by 10%, the borrowing rate grows as well to the 17% neighborhood. The implication is only the weakest dividend future makes this cheap money for the seller. Would you bet that SCOM's dividends will grow or shrink?
The Verdict
Nobody signs a deal to lose. The verdict rests on whether each walked away with the best deal its position allowed. In one of its positions, the government took an advance on the dividends of its remaining 20% stake in SCOM, allowing VKL to collect those future dividends. If the goal was to raise funds by borrowing, GoK had several options, each cheaper than this advance. That week, its own auctions sold one-year bills at 9.4% and twenty-year bonds at 13.6%.² The dividend advance raised KES 40.2 billion at about 16%; four days after signing, the bond market handed the Treasury KES 47.1 billion at 13.3–13.6%, with money left over on the table.² Nobody can argue the market lacked the appetite or the amount. Besides that, Kenya's blue-chip companies were borrowing five-year shillings at between 10.4 and 11.8%. The plainest measure of the gap is Safaricom selling five-year notes at 10.4% in an offer that shut on 5th December.³ Lenders and the government both put a price on Safaricom's cash that week, lenders by funding the company for five years at 10.4%, the government by selling the advance on its dividends at about 16. That's how we know the government sold cheap.
Safaricom was more valuable to the buyer than 34. While the evaluation was from an independent party, the seller underused its discretion to override the valuation.⁴ The seller could have asked for the assumptions to be revisited, commissioned a second opinion, or simply negotiated boldly. In so doing, the seller would have discovered that priced on the profits Safaricom was heading towards, the share was worth 2.66 × 14 = 37.2; range 30–44, mid ≈ 37View full workings (B) →. Weeks before the sale was signed, the company's board had approved half year accounts showing earnings per share up 53%.⁵ Double that half-year and value the share the way the market was already valuing it, at about 16 times earnings, and the price comes to the high thirties. These earnings were a sign that the company's value would only go higher, that's what the arithmetic said. And the market went on to agree. On the day of the agreement, Safaricom's share traded at 28. Seven months later, the price had risen to trade near 35, above the seller's price. One might argue that this rise was driven by the sale as the market moved to price it. However, prices at the NSE moved at around the same rate: the all-share index rose roughly a quarter over those months, KCB from about 66 to 83, Equity from about 67 to 87. Deal or not, Safaricom would have moved the same way. The market and the calculation put Safaricom above 34, the seller took 34.
What the seller gave up on price, the buyer collected, and then some. VKL did what most investors wish they could do, borrow cheap and invest in assets that pay dear. Alongside the shares, VKL bought the dividends to the 20% the government kept, KES 40.2 billion paid now to collect KES 55.7 billion as those dividends come in, in about two to three years.¹ Read like any loan, money out, more money back and time, that works out to a return in the mid-teens. The money to make the purchase was borrowed at an undisclosed rate, but shilling money that season went for around 10 to 12% — the same five-year rate Safaricom's own notes had just carried. Against a cost in that neighborhood, a mid-teens return is money earned on money borrowed.
The share expenditure does not pay for itself, at least not right away, but it bought a growing asset. VKL borrowed at 5.1% earnings yield vs ~8% borrowed; crosses over in ~3 yrs on the 53% EPS growthView full workings (D) → to get 15% of Safaricom. On its first-year profits, the shares return about 5% in earnings and 6% in cash dividends. This return is based on the buyer's price of 34. On the first year's numbers alone, then, the asset does not cover the interest on the money that bought it. But that's hardly a problem, because for one, the interest never grows, but the earnings do. Safaricom's earnings per share rose 53% in the last half year.⁵ At that rate, a return of 5% that grows will pass a fixed cost of 8% within three years. Every year after, the shares earn more than the loan costs. If unconvinced by growth, asset appreciation and value offer a similar case. The buyer quickly gained an edge. Shares bought at 34 are now trading close to 35. In just seven months, they gained value while the debt remained unchanged. Cheap fixed money against a rising asset does not need to pay off in year one; it only needs the asset to keep rising.
A deal is judged against what each side could have done instead. By that measure, GoK left money on the table twice. It borrowed against its own dividends at a mid-teens rate when its bonds cost less. And it sold its shares for 34 when the company's profits pointed nearer to 37. VKL made neither mistake, it paid with borrowed money that earns more than it cost. The distinction arrived on the pricing of the shares. The government priced Safaricom on what it had been; VKL priced it on what it would become, and the value was in the future. A share is a claim on a company's future, not a receipt for its past. The side that remembered this got the better deal.
Notes and Sources
- Vodacom Group Limited. (2025). Acquisition of a further 20% interest in Safaricom PLC [SENS announcement, 4 December]. https://vodacom.com/pdf/sens/2025/acquisition-of-a-further-20-interest-in-safaricom-plc.pdf
- Central Bank of Kenya. (2025). Treasury bond auction results: SDB1/2011/030 and FXD1/2021/025, dated 8 December 2025; Treasury bill auction results, dated 15 December 2025. https://www.centralbank.go.ke/uploads/historical_treasury_bond_results/1478101021_RESULTS%20SDB1-2011-030%20AND%20FXD1-2021-025%20DATED%2008-12-2025.xlsx.pdf; https://www.centralbank.go.ke/uploads/91_day_historical_treasury_bill_results/1154080067_RESULTS%202660-091%202634-182%202589-364%20DATED%2015-12-2025.pdf
- Safaricom PLC. (2025). Pricing supplement: fixed rate green notes, Issue No. SCOM-FXD/T01/2025/05 [5-year, 10.40% per annum, dated 25 November 2025]; East African Breweries PLC. (2025). Medium-term note programme: Tranche 1 offer results [5-year, 11.80%]. https://www.safaricom.co.ke/images/Downloads/Safaricom-2025-MTN-Programme-Pricing-Supplement.pdf; https://www.eabl.com/~/media/Files/E/eabl/documents/investor/mtn-programme/2025/eabl-2025-mtn-programme-public-announcement-on-offer-results.pdf
- National Assembly of Kenya. (2026, March 11). Official report (Hansard), Wednesday, 11 March 2026. Parliament of Kenya. https://www.parliament.go.ke/sites/default/files/2026-03/The%20Hansard%20-%20Wednesday,%2011%20March%202026%20(P).pdf
- Safaricom PLC. (2025). Unaudited results for the six months ended 30 September 2025 [HY26 earnings booklet]. https://www.safaricom.co.ke/images/Downloads/HY26-Earnings-Booklet.pdf
Figures for share prices and market indices are stated as of the dates given and verified against exchange and issuer data. Valuations of ~KES 13, ~KES 37, and the implied borrowing rates are the author's own computations; workings appear in the appendix.
Appendix: The Workings
Every figure the article rounds is shown here in full. Share earnings are per share; the government's retained 20% stake is 8.0 billion shares (a fifth of Safaricom's 40.1 billion).
A. What KES 34 Is Worth on Three Lenses
Each measure asks a different question, so each returns a different number. All use FY2025 figures known at the deal: earnings per share KES 1.74 (three-year average 1.53), book value per share KES 4.61.
| Lens | Working | Result |
|---|---|---|
| Graham number | √(22.5 × 1.53 × 4.61) | ≈ 12.6 (about 13) |
| Price-to-earnings | 34 ÷ 1.74 | 19.5× earnings |
| Dividend yield | 2.00 ÷ 34 | 5.9% |
| Price-to-book | 34 ÷ 4.61 | 7.4× book |
The Graham number uses the three-year average earnings, as Graham's method intends; on the latest year's earnings it is about 13.4. Either way the deal price sits well above it, which is what a deep-value screen is built to flag.
B. The Share Priced on the Future — How KES 37 Is Reached
The half-year to September 2025 earned KES 1.07 per share, up 53% on the year before (0.70 → 1.07 = +52.9%). Group net income rose 22.6%; the sharper 53% is profit attributable to shareholders once the minority share of Ethiopia's losses is stripped out — the per-share figure that governs the share's value. A full year is estimated two ways: simply doubling the half (2.14), or applying the prior year's actual half-to-full split, in which the first half was 40% of the total (2.66). The market was paying between 14 and 16.4 times earnings around the deal. The value is the earnings times the multiple:
| Forward earnings | × 14 | × 16.4 |
|---|---|---|
| 2.14 (half doubled) | 30.0 | 35.1 |
| 2.66 (seasonal split) | 37.2 | 43.6 |
The range is 30 to 44; the middle sits near 37. The check on it is what actually happened: the year closed at KES 2.39 earned per share, and at the ~14.6 multiple the market was paying by mid-2026 that is 34.9 — the price the share reached on its own.
C. The Dividend Advance Read as a Loan
VKL paid KES 40.2 billion for up to KES 55.7 billion of dividends on the retained 8.0-billion-share stake. The rate that makes those equal depends on how fast the dividends arrive, which depends on the dividend per share.
| Dividend case | Annual to VKL | Cleared in | Implied rate |
|---|---|---|---|
| Falls to 1.20 (the 2024–25 level) | KES 9.6bn | ~6 years | 10.5% |
| Holds at 2.00 (FY2026) | KES 16.0bn | ~3.5 years | 16.0% |
| Grows 10% a year | rising | ~3 years | 16.9% |
Against a shilling loan costing around 10–12% — the rate Kenya's strongest names paid for five-year money that season, Safaricom itself at 10.40% and East African Breweries at 11.80% — the steady and growing cases return more than the money cost; only a lasting fall in the dividend closes the gap.
D. The Shares Read as a Leveraged Holding
At 34, the shares return about 5% in earnings on deal-date profits (1.74 ÷ 34 ≈ 5.1%), rising to about 7% on the realised year (2.39 ÷ 34 ≈ 7.0%), and 5.9% in cash dividends, against roughly 8% borrowed. The earnings yield sits below the interest in year one. It does not stay there: on the earnings growth the company was posting, the yield passes 8% within about three years, on the flat assumption that the market keeps paying a mid-teens multiple. The growth assumption is the article's one forward wager; the crossover is sooner if growth is faster, later if slower, and does not arrive at all if earnings stall.
E. The Mandatory-Offer Cost Avoided (Context)
Had the buyer been required to offer 34 to the public's 25% as well — 10.0 billion shares — that offer would have totalled about KES 340 billion. It was not required. This figure is our arithmetic on the public float; it is context for the size of the transaction, not part of the verdict.
Disclaimer: This article is for educational purposes only and does not constitute financial advice. Investing in securities involves risk, including the possible loss of principal. Past performance does not guarantee future results. Readers should conduct their own research and consult with a qualified financial advisor before making investment decisions.