In March, Centum told the market to expect a modest gain. In September, the cost scorecard showed a Sh301 million hole. The board still called the era legendary. I am a long-time shareholder. That ovation was not for us.
I have held Centum for years. I funded the long bets. I sat through the NAV sermons. I watched the share price live at a discount so wide it became a standing joke. On 12 March 2026, James Mworia signed a public announcement completing the sale of Centum’s entire equity stake in Bakki Holdco Limited, the vehicle that held the residual interest in Sidian Bank. The language was triumphant. “Successful completion.” “Important milestone.” Liquidity strengthened. Capital to be reallocated toward new growth. Then the line that should now be framed and hung in the boardroom: The company expected the sale to result in a modest financial gain relative to the previously reported carrying value of Bakki Holdco in Centum’s books.
Carrying value. Not original cost. Not cash-on-cash. The written-down number already sitting in the accounts.
On 9 September, Business Daily reported what that milestone looks like when you measure it the way an owner measures it. Total proceeds from the Sidian disposals: Sh4.469 billion. Original investment cost: Sh4.77 billion. Shortfall: Sh301 million. Recovery: 93.7 percent of historical cost, before counting whatever thin dividends the bank ever sent upstairs. A modest gain against book. A loss against the money we actually put in. That is not a rounding difference. That is the whole method. Write the asset down over the ugly years, sell it, declare victory versus the new lower number, and hope nobody opens the old file. James Mworia signed that March notice. On 7 September he left to become founding CEO of the National Infrastructure Fund. The board called his tenure a “legendary era.”
Legendary for whom?
This was not a distressed leftover: Sidian was not some rotting microfinance remnant they were dumping to stop the bleeding. By the time Centum finished selling, the Central Bank had already upgraded the lender to mid-tier status, in September 2025. FY2025 profit jumped about sixfold to Sh1.73 billion, from Sh287 million. Deposits rose about 63 percent to Sh72.3 billion. Public-sector mandates — Nairobi County, Social Health Authority, housing levy collections — were flooding the balance sheet with cheap deposits. Those deposits were being parked in government paper. The profit line exploded. Into that tape, Centum sold. Then it told the market to expect a modest gain versus carrying value. Then the cost scorecard came out Sh301 million light.
That is the argument. Not mood. The two numbers.
The last trade was an allocation failure
An investment company exists to do three things well: pick assets, hold them through the ugly years if the thesis is intact, and sell them when the market will pay for the work. Centum did the first two on Sidian. It failed the third. The position began in 2001, when the lender was still K-Rep. Centum later built a controlling stake, including the 2014 buy-up that took it to about 67.5 percent, and later as high as 83.43 percent. It injected capital through the rate-cap years, the Imperial and Chase collapses, Covid, and the long stretch when Sidian was subscale and hungry for rights issues. Shareholders funded that hold. That is what “patient capital” is supposed to mean.
Patience is not a virtue if you sell the harvest before it is weighed, then score the sale against a written-down book. In June 2022, Access Bank agreed to pay Sh4.3 billion for the controlling stake. That deal died in January 2023. Deals collapse. What you do next is the test. What they did next was sell the bank in pieces to local buyers — Pioneer General, Wizpro, Afram, then the last Bakki Holdco slice in March 2026. Centum held 50 percent of Bakki. Bakki owned 27.2 percent of Sidian. That last notice did not name the buyer and did not name the price. It named a “modest financial gain” versus carrying value. Business Daily has since put the final 14.63 percent slice at Sh1.2 billion and the whole exit at Sh4.469 billion.
Piecemeal exits can be dressed up as “capital recycling.” They are much harder to defend when the franchise is accelerating, listed Kenyan banks are being re-rated on earnings and dividends, and the final score versus original cost is still a Sh301 million hole. If buyers wanted Kenyan banks, why did a multi-year controlling hold end below cost?
“Mature asset” is the phrase used when the exit is inconvenient
The March notice said the divestment would strengthen liquidity and reallocate capital toward new opportunities. Real estate. REITs. The next story. Sidian was called mature at the exact moment it stopped looking mature. A mid-tier bank with surging deposits and a newly rebuilt public-sector franchise is not a completed trade. It is a scarcity asset. You do not have to love every line of Sidian’s book — loan growth lagged the deposit explosion; a lot of the new profit came from Treasury paper — to see the point. The market was paying up for Kenyan bank earnings. Demand for acquisitions had not vanished. Centum still took cash and walked.
That is an asset-allocation decision. It is not an act of God.
Mworia did not invent this exit by himself. The March announcement went out “BY ORDER OF THE BOARD.” Valuations, buyer selection, the choice to take cash instead of riding the growth, the choice to brief the market on carrying value instead of original cost — those sit with management and the board. If this was a poor deal for owners, both names are on the ticket.
The board gave him a standing ovation
When Mworia left, Centum’s board did not ask the obvious question. It issued gratitude. Assets from Sh4 billion in December 2008 to about Sh46 billion now. Debt retired. A “legendary era of service.” His call to national duty as proof of leadership. Asset growth is not the same thing as owner returns. Centum’s own FY2026 numbers put company NAV at about Sh69.47 a share. The stock closed 9 September around Sh17.85. That is not a rounding error. That is a market that has spent years refusing to believe the NAV will ever become cash in shareholders’ pockets.
Yes, there have been dividends. Small ones. Sh0.32. Then another Sh0.32. This year a proposed ordinary Sh0.42 plus a special Sh0.36 — Sh521 million in total — dressed up as a bumper payout after the exits. Over an 18-year CEO tenure, the company cites about Sh5.8 billion in cumulative declared dividends. Against the capital tied up, the discount at which the stock trades, and the years we sat through write-downs and “recycling,” that is not a distribution record. It is a tip. A board that signs off on “modest gain versus carrying value,” then six months later faces a Sh301 million shortfall versus cost, then blesses the departure without forcing a harder accounting, is not independent. It is complicit.
The timing is the insult
12 March 2026: last Sidian sale completed. Notice signed by Mworia. Modest gain versus book. Months later: Sidian’s growth is no longer a forecast. It is in the published numbers. 7 September 2026: Mworia leaves for the National Infrastructure Fund. The acting CEO, Thomas Omondi-Achola, is the same executive who previously served as Chief Transformation Officer at Sidian. 9 September 2026: the cost scorecard is printed. Sh301 million short.
I am not alleging a plot. I am reading a sequence.
Shareholders funded the ugly years. New owners and a rebuilt deposit base sit with the clean years. Management scored the exit against a carrying value it had already marked down. The man who ran the allocation book takes a state infrastructure job. The board calls it legendary. If Sidian was mature, say so with a price that clears the original cost by a margin that respects the equity risk we carried. If it was not mature, do not sell a rising mid-tier bank, tell the market to expect a modest profit, and then leave the building. Mworia’s last major portfolio act as CEO of Centum was not a harvest for the people who paid for the tree. It was an exit below cost, announced as a gain, followed by applause.
What I want asked at the next AGM
Not poetry. Not “strategy refresh.” These questions:
Against the Sh4.77 billion we put into Sidian, what is the full cash-on-cash record including every dividend received from the bank, every capital injection, every write-down, and every sale tranche? Put carrying value and original cost on the same page.
Why did the 12 March 2026 notice promise a modest gain versus Bakki’s carrying value, and who decided that was the right yardstick for ordinary shareholders?
Why was piecemeal disposal to local buyers superior to holding a growing mid-tier bank through FY2025 and 2026, when the earnings inflection was already visible?
Who signed off on the final valuation and buyer for the Bakki Holdco sale, and why was the buyer undisclosed?
If exits are “unlocking value,” why does CTUM still trade at a roughly three-quarters discount to reported NAV?
What capital-return policy binds the new management, so “recycling” stops meaning “we sold it, we marked it down first, and you can wait”?
I did not buy Centum to fund a museum of almost-exits. I bought it so that when an asset finally worked, the owners who paid for the work would be in the room when the cheque was written.
On Sidian, we were not. We got a press statement about a modest gain. Then we got the cost.
The board can keep using the word legendary. Shareholders should use a different one: accountable.
Sultan
Shareholder, Centum Investment PLC
About Boardlot Africa Research
Boardlot Africa is a premier financial intelligence and corporate governance publication dedicated to unpacking the mechanics of capital, market strategies, and structural shifts across East Africa’s corporate landscape. By bridging the gap between raw economic data and actionable market intelligence, we deliver deep-dive research, independent corporate analysis, and policy insights designed for institutional investors, boardrooms, and sharp market observers.
Get in Touch
Email: boardlot.research@gmail.com
Phone: +254 753 133 901
Substack: Subscribe to Boardlot Africa
X (Twitter): BoardLotSultan
I have a few questions of my own:
Who really are the buyers of Sidian? The actual individuals not some amorphous corporation?
Why would Mworia take such a significant paycut to get into public service, with the very well known attendant risks therein, which risks have crystallized and burnt others before him?
Why did he really exit Centum?
Kudos Bro this is well researched 💯