If two companies deliver the same work but one holds more cash and the other borrows more, how much does a difference in ROE tell us about their core operations?
Return on net operating assets (RNOA) sharpens that question. It separates assets and liabilities used in operations from financial items, then compares after-tax operating profit with the net base used to do the work.
We continue with the hypothetical equipment seller, systems installer, and support provider. Amounts are THB millions. Y1 and Y2 are 12-month periods; Y0 is the balance sheet just before Y1. Classify each item consistently before dividing any figures.
Classify items by their role in the business
Trade receivables, inventory, work with recognized revenue but a conditional payment right, and delivery equipment are operating assets. Trade payables, accrued operating expenses, and customer advances help finance operations and are subtracted to find the net operating base.
Interest-bearing borrowing is financing. Cash classification must fit the business. Here THB 2 million is reserved as operating cash at every period-end, and the rest is a financial asset. That is a case assumption, not a default rule for every company.
NOA = operating assets − operating liabilities, while NFO = financial liabilities − financial assets. We use NFO to mean net financial obligations and require NOA = NFO + equity to reconcile.
A real company may have leases, taxes payable, investments, pensions, or other items needing more information. The simplified example leaves them out to show the mechanism; an account name alone does not always dictate its classification.
Recast the same statements without changing totals
| Item | Y0 | Y1 | Y2 |
|---|---|---|---|
| Operating cash | 2 | 2 | 2 |
| Trade receivables | 12 | 16 | 22 |
| Inventory | 8 | 10 | 13 |
| Contract assets | 4 | 6 | 9 |
| Prepaid expenses | 1 | 1 | 1 |
| Net property, plant, and equipment | 25 | 27 | 30 |
| Total operating assets | 52 | 62 | 77 |
| Less payables, accrued expenses, and customer advances | 15 | 20 | 25 |
| NOA | 37 | 42 | 52 |
| Item | Y0 | Y1 | Y2 |
|---|---|---|---|
| Interest-bearing debt | 15 | 18 | 22 |
| Less cash classified as a financial asset | 3 | 2 | 4 |
| NFO | 12 | 16 | 18 |
| Equity | 25 | 26 | 34 |
| NFO + equity | 37 | 42 | 52 |
NOA equals NFO plus equity on every balance-sheet date. The totals for assets, liabilities, and equity have not changed; only the analytical classification has.
Which operating assets and liabilities raised closing NOA?
Year-end components, THB millions
Separate panel scale: −30 to +30 THB million
Operating assets positive; subtracted operating liabilities negative
Operating cash
Trade receivables
Equipment inventory
Contract assets
Prepaid expenses
Net property, plant, and equipment
Less operating liabilities
Total net operating base at year-end
Chart scale: 0 to THB 60 million
Closing NOA
Choose profit that matches the operating base
With NOA as denominator, the profit numerator should exclude financing effects. NOPAT means net operating profit after tax. In this simplified model, it is EBIT × (1 − assumed tax rate).
Thus Y1 is 15 × (1 − 20%) = THB 12 million, and Y2 is 18 × (1 − 20%) = THB 14.4 million. The 20% rate is only an assumption. Interest is assumed fully tax-deductible, with no financial income or other tax items.
NOPAT is neither net profit nor cash received from work. In Y2, THB 12.4 million net profit equals THB 14.4 million NOPAT less THB 2 million after-tax interest. CFO of THB 11.4 million also reflects changes in receivables, inventory, and other operating balances.
Real statements require reconciliation of operating and financing taxes. Do not multiply EBIT by a rate seen in the news or used by another business without checking the facts.
Calculate RNOA, then split margin from turnover
RNOA is period NOPAT divided by average NOA. Y1 average NOA is (37 + 42) ÷ 2 = THB 39.5 million; Y2 is (42 + 52) ÷ 2 = THB 47 million.
| Item | Y1 | Y2 |
|---|---|---|
| Revenue | 100.00 | 125.00 |
| NOPAT | 12.00 | 14.40 |
| Average NOA | 39.50 | 47.00 |
| NOPAT ÷ revenue | 12.00% | 11.52% |
| Revenue ÷ average NOA | 2.5316 times | 2.6596 times |
| RNOA | 30.38% | 30.64% |
NOPAT margin falls, but NOA turnover rises enough for RNOA to improve slightly. We still need to identify what work grew revenue and what changed in NOA. The product is a mathematical relationship, not proof of the cause.
If average NOA is zero, the return is undefined. If negative, inspect the operating asset/liability structure rather than ranking the percentage. A business that collects substantial advances may have a very different net base from one giving customers long credit.
NOA turnover shows revenue generated per baht of net operating base, not machine utilization or employee hours directly. High RNOA also does not establish value creation until compared with a required return on a consistent base.
Why does a different cash classification change RNOA?
If all cash is treated as a financial asset, NOA falls to THB 35, 40, and 50 million at Y0/Y1/Y2. Y1 and Y2 averages become THB 37.5 and 45 million. Dividing the same THB 12 and THB 14.4 million NOPAT gives 32% RNOA in both years.
This is a sensitivity test of classification, not an overnight improvement in operations. A smaller denominator changes the result. Compare RNOA using the same classification policy and disclose large classification choices.
The main series consistently uses THB 2 million operating cash. The 32% figure is a separate test. Appropriate operating cash for a real business depends on receipt/payment cycles and work risks.
Connect back to ROE without substituting D/E for FLEV
The Nissim and Penman formula links common-shareholder return to operating and financing returns. This case has no preferred shares, noncontrolling interests, or other comprehensive income items.
ROE = RNOA + FLEV × (RNOA − NBC)
FLEV is average NFO divided by average common equity. NBC is after-tax net financial expense divided by average NFO, not the contractual loan interest rate. In Y2, FLEV = 17 ÷ 30 = 0.5667 times, and NBC = 2 ÷ 17 = 11.7647%.
With unrounded figures, ROE is about 30.6383% + 0.5667 × (30.6383% − 11.7647%) = 41.3333%, matching net profit 12.4 divided by average equity of THB 30 million. The financing contribution is about 10.70 percentage points.
The formula requires matching after-tax profit and average bases. With zero average NFO, NBC is undefined. A net-financial-asset company should separately explain returns on those assets. A positive historical spread does not guarantee that more borrowing will help in the future. The source framework is Ratio Analysis and Equity Valuation: From Research to Practice.
Bring NOA back to work that can be improved
For this hypothetical company, NOA rises from THB 42 million to THB 52 million. Some is in receivables, contract assets, and inventory; some in net fixed assets. Separate investment supporting growth from balances stuck because of problems, using aging and work documents rather than totals alone.
Software and SI teams should separate recognized revenue still subject to billing conditions from unconditional receivables, and both from product development costs that need their own asset-recognition assessment under the applicable accounting policy. Calling all of it IP or WIP does not tell us when cash will arrive.
A hypothetical factory might separate inventory by age, work in progress, and new equipment not yet at full capacity. Pushing NOA down without checking delivery capacity can interrupt work. The aim is an explainable invested base with returns appropriate to risk.
Before using RNOA in a management report, have accounting confirm one classification schedule. Ask work owners to explain the largest increase in NOA and name the conditions under which it will become revenue or cash.
Source framework and scope of the example
The operating-base, NOPAT, and turnover concepts are applied to invented statements. The common-shareholder return identity draws on the linked Nissim and Penman work. The amounts, THB 2 million operating-cash policy, and team applications were created for this series; they are not company data or industry benchmarks from that research.
