NSE cross-sector casebook: seven businesses, seven cash questions

Revised practical article; source details and figures are dated to their cited filing or event, with current evidence reviewed through 24 September 2026.

An industrial-company formula can be useful for a Kenyan listed business, but only after we know what its assets and liabilities do. A bank’s customer deposits fund lending. An insurer’s premiums come with future claims. A farm may harvest seasonally and carry biological assets. A utility spends before a plant produces electricity. A telecom network needs continuing capital and spectrum. A manufacturer buys inputs and funds inventory. A property developer may have land, unsold units and customer instalments, but cash tied up in a project can remain unavailable for years.

This casebook uses seven NSE research anchors to ask the same first question in different ways: what cash can the owners ultimately claim after the business funds the operating system that earns it?

The examples above are chosen because their operating models differ, not because they have similar valuation multiples. Equity’s 2025 Integrated Report, Jubilee’s 2025 Annual Integrated Report, KenGen’s FY2025 Integrated Report, Kakuzi’s 2025 annual report, BAT Kenya’s 2025 report, Home Afrika’s 2025 reporting, and Safaricom’s reporting catalogue provide dated primary starting points [1]–[7]. Each article calculation should preserve the fiscal period, entity and unit in the source.

Three quick illustrations

Agriculture: profit can lead the cash by a season

A grower can record value movements in trees, fruit, or other biological assets before receiving sale proceeds. Harvest volume, grade, export acceptance, prices, packing costs, transport, and buyer-payment timing all affect cash. For Kakuzi, an analyst would compare crop volume and prices with the cash-flow statement, read the biological-assets and bearer-plant notes, then trace investment in plantations and packhouse capacity. The company’s official report record includes its FY2025 annual report and FY2026 half-year publication [4]. The analyst still needs to open the filed reports and link any number used into an evidence ledger.

Telecom: infrastructure must carry the forecast traffic

A subscriber count does not tell us whether the network has enough capacity, how often equipment must be refreshed, or what power and lease commitments apply. For Safaricom, follow mobile connectivity and M-PESA separately, then reconcile capital spending, leases, segment results, financing, and cash generation to the group statements. Its reporting catalogue is the primary starting point for dated filings and results [5]. If a service grows faster than the network can carry it, expansion spending may precede revenue. If capex is cut, near-term cash may look better while customer experience or future capacity deteriorates.

Property: deposits can arrive before an earnings story is complete

A developer might receive deposits before construction is complete, yet still need cash for contractors, approvals, infrastructure, marketing, debt interest, and handover. A buyer’s instalment is both a receipt and evidence that the seller owes a completed unit or other contractual performance. The model needs project-by-project completion estimates, customer refund rights, and the timing of remaining costs. Home Afrika publishes annual reports and financial publications by reporting year, which supports a dated record rather than reliance on a project headline [7].

Why one multiple is not enough

A price-to-earnings ratio uses equity market value against profit attributable to equity owners. A price-to-book ratio is often useful where a regulated balance sheet is central to earnings. An enterprise-value multiple can work for a nonfinancial operating company when debt and cash are consistently defined. None of those measures makes unlike businesses comparable automatically.

Ohlson’s peer-reviewed valuation framework connects book value, earnings and dividends under an explicit model [8]. Its practical lesson here is not that one formula fits every sector. It is that a valuation depends on future earnings and the assets and capital supporting them. For banks, fair-value information can affect how investors interpret the portfolio and the equity claim [9]. For retailers and manufacturers, working capital absorbs or releases cash, and its economic value depends on expected sales, credit risk and financial constraints [10]. For energy and property firms, replacement and project investment shape the length and scale of future cash generation [11].

Build one method around each issuer’s economics, then use a cross-sector table only to organize the research. For each name, record the question the model can answer and the disclosures it lacks. Do not force a value estimate when the necessary operating or capital information is missing.

A reporting rule for this casebook

A useful NSE comparison table records the report period, publication date, currency and unit; source-file name and page; consolidated or parent scope; reported earnings; cash from operations; capital expenditure; debt and lease claims; shares; dividend declaration and payment date; and unresolved items. Market prices need a separate dated market-source observation. A return calculation should not pair a price from one date with an earnings period from another without labeling the mismatch.

The previous Kenyan Equities and Company Analysis project already has four final workbook roles: screening, statements and valuation, portfolio construction, and compounding. The NSE Dividend Growth workbook separates contributions and income scenarios. Those are helpful companion products, but this casebook’s central task is to make each owner-earnings bridge traceable to its issuer-specific sources and investment needs.

Seven issuer cards: what to calculate from the current reports

Equity Group: earnings growth still has to pass the capital test

Equity Group's FY2025 results report KES 75.5 billion of profit after tax, total assets of KES 1.97 trillion, customer deposits of KES 1.46 trillion and net loans of KES 882.5 billion. The announcement also reports a 1.7% cost of risk and 67.7% NPL coverage, and proposes KES 21.7 billion of dividends, or KES 5.75 per share [12]. The proposed cash dividend is approximately 28.7% of reported group profit after tax. That is a descriptive ratio, not proof of capital headroom.

Build a bank model from average loan and deposit balances, net interest income, fees, credit costs, regional subsidiary contributions, regulatory capital and risk-weighted assets. Separate Equity Bank Kenya from the regional banks and other group activities. Loan growth is not free: it may increase risk-weighted assets and require retained capital. Check whether the strong reported earnings are supported by cashable recurring income, and test a higher credit cost before carrying the proposed payout forward.

Jubilee Holdings: premium and deposit contributions are not earned profit

Jubilee's FY2025 announcement reports gross written premiums and deposit administration contributions of KES 62.4 billion, net profit of KES 5.6 billion and profit before tax of KES 7.2 billion; its H1 2026 update later reports profit before tax of KES 4.4 billion [13]. These are issuer-reported measures for different periods and must not be blended into a single annual run-rate.

Build by insurance line and regulated entity: insurance service result, claims, acquisition expense, reserve development, investment result, reinsurance, IFRS 17 balances, capital and distributions from subsidiaries to the parent. Compare the dividend with after-tax earnings and capital actually available at the paying entity. Ask how much of profit came from underwriting and how much from investment or finance movements. An increase in premiums can mean more future service obligations as well as a larger customer base.

KenGen: profit competes with long-lived plant spending

KenGen's FY2025 integrated report records KES 10.481 billion of profit after tax, KES 17.733 billion of capital investment and KES 30.124 billion of cash and cash equivalents, all in the report's KES-million presentation [3, 14]. Capital investment was about 1.69 times reported profit after tax. This does not mean all investment is maintenance or that none can be deferred; it shows why the dividend cannot be assessed from profit alone.

Trace generation, availability, tariffs, fuel and geothermal costs, receivables from the off-taker, project capex, debt, interest and project commissioning. Separate refurbishment of existing plants from new generation and transmission-related projects. Model delayed payments and project overruns beside generation and debt service. Cash held today may support capital commitments already signed and should not be counted again as an uncommitted shareholder reserve.

Kakuzi: use crop-level cash timing, not one annual profit figure

Kakuzi reported FY2025 revenue of KES 5.369 billion and profit before tax of KES 568 million, versus a KES 167 million pre-tax loss in FY2024; the company also reports that 2025 avocado profit was KES 709 million and macadamia profit KES 365 million [4, 15]. The reported FY2025 turnaround is real as an accounting result, but segment profits are not cash collected and do not add directly to a distributable number after group costs, tax, working capital and renewal.

Build one schedule per crop and operating line. Include planted hectares, tree age, yield, grade, export volume, selling price and currency, harvest timing, packhouse and transport cost, buyer acceptance, collection lag, orchard renewal and biological-asset accounting. The chairman's report describes logistics disruption, pest and disease pressure, lower European prices and a developing route-to-market strategy [15]. These are cash and margin drivers, not background color. Stress a poor-quality shipment and delayed customer remittance, then show whether the dividend is supported from harvest proceeds or from cash accumulated in another season.

Safaricom: split recurring telecom cash from network and market-entry investment

Start with the FY2026 results booklet and the segment and cash-flow notes [5, 16]. Keep Kenya and Ethiopia separate where disclosure allows. Reconcile service revenue, mobile money and data, voice and other income, operating cost, EBITDA, cash from operations, capex, leases, tax and debt. Customer growth is a volume measure; it is not a cash measure until activity is monetized, service costs and partner shares are paid, and customers or agents settle.

For the mature Kenyan business, estimate network renewal and expansion by radio access, core systems, fibre, towers, spectrum, IT and mobile-money systems. For Ethiopia, separate the rollout and customer-acquisition phase from the established operation; currency translation, network investment and startup losses can dominate group trends. Test capex cuts against coverage, capacity, quality and future traffic. A low-capex year could temporarily lift free cash while delaying replacement. Mobile-money float and customer funds are not company cash available for dividends simply because the platform processes large transaction values.

BAT Kenya: taxes, volume pressure and distributions must be reconciled

BAT Kenya's audited FY2025 result reports net revenue of KES 23.192 billion, down 10%; operating profit of KES 7.475 billion; profit before tax of KES 7.671 billion; net cash from operating activities of KES 6.642 billion; and total dividend of KES 70 per share [6, 17]. Its reported 18% growth in pre-tax profit occurred alongside declining net revenue, lower operating cost and a shift in finance result. The investor should separate operating performance from currency or treasury effects and identify whether cost reductions are repeatable without damaging the revenue base.

Rebuild gross sales, excise and VAT, net revenue, operating cost, working capital, tax, capex and cash distributions. Excise and VAT are not operating profit, but their timing and compliance obligations matter to cash. Compare volume, price, product mix and domestic versus export sales; model the effect of continued volume decline and the contribution of alternative products separately. The KES 7.0 billion dividend is higher than KES 5.246 billion of FY2025 profit after tax, so the analysis should ask whether it is supported by retained cash, prior-year earnings or another capital source. The cash-flow statement reports KES 6.642 billion of operating cash before investing and financing, but total owner cash still depends on capital needs and the timing of the payout [17].

Home Afrika: a signed sale or presale is not yet distributable cash

Read Home Afrika's 2025 integrated report, consolidated statements, project schedule and 2026 half-year publication as dated documents [7, 18, 19]. Build a separate cash bridge for each development: opening land and work in progress; completed units; signed reservations and sales; customer deposits collected; remaining construction and infrastructure spend; borrowing and interest; expected handover; refunds or cancellations; and final collection. Determine which entity owns each parcel, project, receivable and debt. A group-level sales headline can hide cash trapped in a subsidiary or a project with a large completion obligation.

For each unit or phase, compare expected net proceeds with cost to complete and the time until cash is unrestricted. Check land title, approvals, contractor claims, security interests, customer rights, service charges and any presale escrow restrictions. An unsold parcel may be valuable but does not automatically fund payroll or a shareholder dividend. Likewise, customer deposits can be a useful financing source but remain linked to delivery and refund obligations. Use the half-year report to see whether promised completion milestones turned into handovers and collections, not just whether reported revenue increased.

Put the companies on one page without forcing a common formula

A cross-sector table should compare the questions, evidence and unresolved risks—not multiply unlike earnings by one shared ratio.

The first output is therefore not a league table saying which issuer has the most “owner earnings.” It is seven dated bridges with source links and a clear record of what remains unmeasured. Then the analyst can compare valuation and expected return within the right capital structure. A bank may be best evaluated with a residual-income or dividend-capacity model, a developer with project-level net cash realization, and a utility with operating assets and project funding. Ohlson's framework and the peer-reviewed evidence on working capital, bank valuation and replacement capital support careful method selection rather than a single cross-sector shortcut [8–11].

How to use this casebook in an investment meeting

For each company, one participant should present filed facts, a second should challenge the cash bridge, and a third should test the downside scenario. End with a short decision record: what is repeatable, what must be reinvested or retained, which claim ranks ahead of ordinary shareholders, what the share price implies, and which new evidence would change the view.

An analyst should be willing to say “not yet estimable” for a specific component while still completing the rest of the work. For example, Home Afrika's public numbers may support a project cash schedule but not a reliable collection date; Kakuzi's crop cycle may be visible while orchard renewal cost is still poorly split; Safaricom's total capex may be reported while a clean maintenance allocation is not. Name the missing source and its likely impact. The workbook can then preserve a useful range and a research task rather than inventing precision.

A repeatable issuer-review workflow

The casebook can be used as a quarterly research process rather than a one-time comparison. For each issuer, follow the same sequence while allowing the economic bridge to differ:

1. Freeze the information set. Save the report and announcement, publication date, fiscal period, entity scope, currency and unit. Do not mix a later press release with an earlier market-price date in a backtest.
2. Transcribe reported facts. Enter statement and note values with the page number and original label. Reconcile the cash, debt, equity, earnings and share-count roll-forwards where possible.
3. Build the sector bridge. Use retained capital for a bank; claims and solvency for an insurer; asset renewal and projects for a utility; crops and harvest cash for agribusiness; network renewal for telecom; taxes and volume for manufacturing; and project cost-to-complete for property.
4. Mark assumptions and gaps. Label each number as filed, calculated, analyst-estimated, or unresolved. Use a range where evidence supports a range; leave a component unavailable when no evidence supports even a range.
5. Run one joined downside. Choose a plausible operating story and connect it to margin, working capital, capex, funding, capital retention and per-share cash.
6. Write a decision note. State valuation method, current price date, required return or margin of safety, key risk, position size or wait condition, and the next filing or event that could change the estimate.
7. Review the prior call. Compare the last forecast with actual cash, capital, projects, claims, volumes and collections. Record what changed and why.

The sequence creates comparability in source quality and review discipline, not in the underlying company formula. A well-prepared research page can show that the greatest uncertainty is disclosed maintenance spending for one company, credit migration for another, and project collection timing for a third. That difference is itself a useful conclusion.

Common analytical errors across sectors

Several mistakes recur even when the companies are very different. First, analysts compare a calendar year with a fiscal year or parent-only results with consolidated results without noting the mismatch. Second, they treat a press-release measure as if it were an audited cash-flow line. Third, they compare a dividend per share with group net income while omitting payout date, tax, new shares or capital requirements. Fourth, they use EBITDA as a proxy for owners' cash before accounting for debt service, working capital, capex and leases.

A fifth error is to call a one-year increase in cash flow recurring. A favorable movement in receivables, inventory or payables can reverse; an asset disposal can create a large one-off receipt; lower maintenance can defer a bill rather than eliminate it. A sixth error is to use segment profit as though it were distributable. Segment results may not include central cost, tax, financing, minority interests, holding-company obligations or cash restrictions.

A practical response is to build a bridge from the measure most often quoted to the measure relevant to owners. For a bank, bridge reported earnings to retained regulatory capital and payout. For Jubilee, bridge insurance service and investment outcomes to parent cash and subsidiary solvency. For KenGen, bridge profit through plant renewal, projects, working capital and debt. For Kakuzi, bridge biological and crop results to harvest receipts and seasonal funding. For Safaricom, bridge service revenue through network and market investment. For BAT Kenya, bridge operating cash after excise, tax, working capital and dividend. For Home Afrika, bridge project revenue through cost to complete, debt and collected cash.

Research quality and source hierarchy

Peer-reviewed research helps discipline the method, but it should not substitute for issuer facts. Ohlson's model makes clear that earnings and book value matter within explicit valuation assumptions [8]. Barth's work shows that disclosures on financial instruments can be relevant to bank valuation [9]. Kieschnick, Laplante and Moussawi study the relationship between working capital management and shareholder wealth [10]. Livdan and Nezlobin examine how non-geometric depreciation changes investment and replacement-cost analysis [11]. These studies provide frameworks and empirical results in defined samples; none supplies a universal owner-earnings answer for the seven Kenyan companies.

For each claim, prefer an audited report, statutory filing, regulatory return or court record where available. Use a company announcement for the date and terms of a proposal, then check the final filed statement for recognition and cash movement. Use a media report to locate a development or understand context, but identify when the underlying primary document has not been reviewed. A chart or table copied into a presentation should keep the source and exact reporting period.

The peer review should ask whether the source supports the specific sentence, not only whether the source is reputable. If a source says that revenue rose, it does not prove a positive cash conversion. If a report gives capital investment, it does not prove all spending was growth. If a court judgment summarizes an administrator's report, distinguish the summary from an audited statement.

Make the next research action visible

Every workbook row marked unresolved should have a named next action and an estimated decision impact. For example: “Request project-by-project maintenance capex; current base estimate is 20% of total capex and value changes materially if replacement spending is higher”; or “Obtain statutory capital by insurance subsidiary; parent payout model is incomplete without it.” The team should set a due date or an event trigger, not leave the cell as a vague open question.

A comparison across sectors becomes more useful when it reveals where better evidence would change a decision. It should show the owner where cash is earned, where it must stay, when it may become available, and what could prevent it from arriving. The casebook is a guide to those questions, not a contest to rank seven unlike businesses on an artificial common metric.

Workbook map

The Project 2 workbook contains a dated evidence matrix for Equity Group, Jubilee, KenGen, Kakuzi, Safaricom, BAT Kenya and Home Afrika. Each row retains its own period, units, measures and source ID; the KenGen capital-investment-to-PAT ratio is calculated from its reported inputs. Evidence Vectors expands populated figures into observation-level records, separates fiscal year-end from publication date (including KenGen's June year-end), fills the missing Equity Group source ID, and leaves Home Afrika's project metrics unpopulated rather than zero. The Profile Stress sheet adds a separate normalized scenario for each of these seven NSE profiles and for NVIDIA and Equinix in the US cases. For example, a crop-yield shock increases seasonal working-capital pressure for Kakuzi, while a customer-collection delay and completion-cost overrun create a funding gap in the fictional Home Afrika profile. Those indices are teaching scenarios, not issuer estimates or a cross-sector ranking. Appendix 04 reproduces all nine assumptions and outputs. The matrix and stress sheet support source selection and issuer-specific questions; neither pretends every issuer already has a complete owner-earnings bridge.

Technical appendix

• Appendix 04 — Listed-company profile stress cases

Technical appendices

The worked calculations and source boundaries below extend this article's examples. They keep assumptions and unresolved inputs visible so readers can reproduce the steps without mistaking an illustration for an issuer or sector estimate.

• Appendix 01 - From the filed statement to an equity claim
• Appendix 03 - Regulated financial companies and sector-specific cash questions

References

[1] Equity Group Holdings PLC, 2025 Integrated Report and Financial Statements, https://equitygroupholdings.com/wp-content/uploads/2026/06/Equity-Group-Holdings-PLC-2025-Integrated-Report-and-Financial-Statements.pdf.

[2] Jubilee Holdings Limited, “Annual Reports,” including 2025 Annual Integrated Report, https://jubileeinsurance.com/group/investor-relations/annual-reports/.

[3] KenGen PLC, Integrated Annual Report and Financial Statements 2025, https://www.kengen.co.ke/images/2025/KenGen_Integrated_Annual_Report__Financial_Statement_2025.pdf.

[4] Kakuzi PLC, “Company Reports,” including the Annual Report and Financial Statements 2025, https://www.kakuzi.co.ke/company-reports.

[5] Safaricom PLC, “Financial Reporting,” https://www.safaricom.co.ke/investor-relations-landing/reports/financial-report/financial-results.

[6] BAT Kenya PLC, Combined Annual and Sustainability Report 2025, https://www.batkenya.com/content/dam/endmarkets/ke/en/download/investors-and-reporting/annual-reports/BATK_Combined_Annual_and_Sustainability_Report_2025.pdf.

[7] Home Afrika Limited, “Annual Reports,” https://homeafrika.com/investor-info/annual-reports.

[8] J. A. Ohlson, “Earnings, Book Values, and Dividends in Equity Valuation,” Contemporary Accounting Research, vol. 11, no. 2, pp. 661–687, 1995, https://doi.org/10.1111/j.1911-3846.1995.tb00461.x.

[9] M. E. Barth, “Fair Value Accounting: Evidence from Investment Securities and the Market Valuation of Banks,” The Accounting Review, vol. 69, no. 1, pp. 1–25, 1994.

[10] R. Kieschnick, M. Laplante, and R. Moussawi, “Working Capital Management and Shareholders’ Wealth,” Review of Finance, vol. 17, no. 5, pp. 1827–1852, 2013, https://doi.org/10.1093/rof/rfs043.

[11] D. Livdan and A. Nezlobin, “Investment, Capital Stock, and Replacement Cost of Assets When Economic Depreciation Is Non-Geometric,” Journal of Financial Economics, vol. 142, no. 3, pp. 1444–1469, 2021, https://doi.org/10.1016/j.jfineco.2021.05.021.

[12] Equity Group Holdings PLC, “FY2025 Results,” 18 March 2026, https://equitygroupholdings.com/equity-group-records-historic-results-reporting-a-55-growth-in-profit-after-tax-of-kshs-75-5-billion-underscoring-the-success-of-its-strategic-transformation/.

[13] Jubilee Holdings Limited, FY2025 results and H1 2026 results, 9 April 2026 and 28 August 2026, https://jubileeinsurance.com/group/newsroom/jubilee-holdings-declares-a-record-dividend-following-18-increase-in-net-profit/ and https://jubileeinsurance.com/group/newsroom/jubilee-holdings-limited-delivers-30-growth-in-h1-profit-before-tax-to-kes-4-4-billion/.

[14] KenGen PLC, Integrated Annual Report and Financial Statements 2025 and official FY2025 results, https://www.kengen.co.ke/investor-relations/financial-information/.

[15] Kakuzi PLC, “Five Year Record,” FY2021–FY2025, https://www.kakuzi.co.ke/five-year-record.

[16] Safaricom PLC, FY2026 Results Booklet, audited results for year ended 31 March 2026, https://www.safaricom.co.ke/images/Downloads/FY26-Earnings-Booklet-2026-7th-May-2026.pdf.

[17] British American Tobacco Kenya PLC, Audited Results for the Year Ended 31 December 2025, https://www.batkenya.com/content/dam/endmarkets/ke/en/download/investors-and-reporting/financial-statements/bat_kenya-2025_full_year_results.pdf.

[18] Home Afrika Limited, Integrated Annual Report and Financial Statements 2025, https://www.homeafrika.com/static/uploads/media/integrated-annual-report-financial-statements-2025-1782322991.pdf.

[19] Home Afrika Limited, Half Year Financial Publication 2026, https://www.homeafrika.com/static/uploads/media/home-afrika-half-the-year-financial-publication-2026-1787922272.pdf.

Companion resources

The workbook and technical appendices are available with research membership. The video production guide maps this article to a planned episode; a published video is not yet available.