Cash flow and earnings quality

In depth

The measures and their conventions

Four measures cover most of what an analyst needs. Each is defined exactly as in the course’s ratio engine, so the lessons, the exercises and the ratios widget agree.

• CFO ÷ net income. How many dollars of operating cash each dollar of profit produced. Meaningless when net income is near zero or negative: then compare the dollar amounts.
• Total accruals = net income − CFO. Positive when profit runs ahead of cash, negative when cash runs ahead.
• Accruals ratio = (net income − CFO) ÷ average total assets. Scaling by assets lets you compare years and companies of different sizes. For 2027 the average is of the opening and closing balance sheets; for 2026 there is no opening balance sheet (Lantern started on 1 March 2026 with nothing), so 2026 uses closing total assets.
• Free cash flow = CFO − capital expenditure, and the FCF margin = FCF ÷ revenue. Some analysts also subtract dividends, or only the capex needed to maintain the existing assets; OpenStax’s accounting text shows both versions. State the definition you use.

Lantern’s accruals ratio was $-12{,}553 \div 93{,}676 = -13.4\%$ in 2026 (closing total assets) and $14{,}090 \div 210{,}779.5 = 6.7\%$ in 2027, where 210,779.5 is the average of $93,676 and $327,883. The 2027 figure is flattered by the size of the balance sheet: $173,379 of it is the cafés’ right-of-use asset, which ASC 842 brought onto the balance sheet in March. Treat a large positive accruals ratio as a prompt to look for the reason, not as a verdict.

A fifth measure looks at investment: capex ÷ depreciation. A business that spends about as much as its depreciation is maintaining its assets; one that spends much more is growing them. Lantern’s 2027 ratio was $54{,}000 \div 19{,}800 = 2.73$, a growth year. A ratio well below 1 year after year can mean assets are wearing out without replacement, which flatters free cash flow for a while.

Estimates inside earnings

Every profit contains estimates. The analyst’s job is to find the large ones and ask how much they could move. Lantern’s 2027 statements have four.

• The contract asset. The law firm’s fee is $40,000 a year in March 2027 dollars, raised by the consumer price index. Lantern estimated 2.8% inflation, giving $40{,}000 \times 1.028 = 41{,}120$ for the first contract year and $41{,}120 \times 9/12 = 30{,}840$ for April–December. Each percentage point of inflation moves 2027 revenue by $40{,}000 \times 0.01 \times 9/12 = 300$ dollars. The estimate is small; the timing and the single client are the real exposure.
• The allowance for doubtful accounts. Set from an aging of catering receivables; a higher loss rate on the over-60-day band would cut profit at once.
• Useful lives. The fit-out is depreciated over the five-year leases and the machines over five years. Longer lives would mean lower depreciation and higher profit, with no change in cash.
• The lease measurement. The right-of-use asset and lease liability ($202,255 at the start) use Coastline’s 7.2% as the discount rate. The straight-line lease cost of $40,000 doesn’t depend on it, but the balance sheet does.

Notice what the list shows: estimates change profit, never cash. That is why comparing profit with cash is the most reliable single test of earnings quality. Cash from operations can be shifted between years too, by delaying payments to suppliers or selling receivables, but it is much harder to create from nothing.

Leases and comparing operating cash flows

Under US GAAP (ASC 842) the payments on an operating lease are operating cash outflows: Lantern’s $40,000 of café rent is inside its CFO. Under IFRS 16 there is no operating lease category for lessees, and the principal part of each lease payment is a financing outflow. And since IAS 7 was amended by IFRS 18, for periods beginning on or after 1 January 2027, a company whose main business is not lending or investing must classify interest paid as financing too; the old choice between operating and financing is gone. On IFRS, then, Lantern’s 2027 operating cash flow would be $40,000 higher: the principal of $40{,}000 - 11{,}124 = 28{,}876$ dollars and the $11,124 of interest would both be financing outflows. Free cash flow would rise by the same amount, though nothing about the cafés’ economics changed.

Reinvestment over reported profit: Amazon, 2012–2014

The opposite pattern to Lantern’s is a business whose reported profit looks poor while its operations produce a great deal of cash. Amazon’s filings for 2012–2014 are a well-known example.

Source: Amazon.com, Inc., Form 10-K for 2014: statements of operations and cash flows, and Item 6 (free cash flow, footnote 2). The 2014 10-K defines free cash flow as net cash from operating activities less those purchases.

In 2014 Amazon reported a net loss of $241 million and operating cash of $6,842 million. The gap was $7,083 million. Depreciation and amortisation ($4,746 million) and stock-based compensation ($1,497 million), both noncash, explain most of it; working capital released a further $974 million, because payables grew faster than inventories plus receivables. Amazon’s operating working capital (receivables plus inventories minus payables) was negative: at the end of 2014 it owed suppliers $16,459 million, more than its inventories ($8,299 million) and receivables ($5,612 million) together, a net −$2,548 million. Customers paid before Amazon paid its suppliers, so growth released cash instead of absorbing it. (Its total working capital, all current assets less all current liabilities, was positive: $31,327 million against $28,089 million.) Free cash flow stayed positive while profit was close to zero.

The lesson cuts both ways. Low profit with strong cash flow can be a sign of conservative or investment-heavy accounting, not weakness. But Amazon’s model depends on suppliers’ terms; a business with Lantern’s structure, where a large client pays a year after the first event, sits on the other side of that trade.

Aggressive accounting: Enron’s restatement

On 8 November 2001 Enron filed a Form 8-K saying it would restate its financial statements for 1997 to 2001. Entities it had kept off its balance sheet should have been consolidated: Chewco and JEDI from November 1997, and a subsidiary of LJM1, a related-party partnership, from 1999. The filing’s Table 1 shows the effect. The restatement (mainly consolidation, plus earlier proposed audit adjustments and reclassifications) cut net income by $96 million (1997), $113 million (1998), $250 million (1999) and $132 million (2000); consolidation alone accounted for $45, $107, $248 and $99 million of that. Consolidating JEDI and Chewco raised debt by $711, $561, $685 and $628 million in those years. For 2000, reported net income of $979 million became $847 million.

The Board’s special investigative committee reported in February 2002 (the Powers Report). Its summary found that many of the most significant transactions were designed to produce favourable financial-statement results rather than real economic objectives or a transfer of risk, and that the hedging transactions with those entities led Enron to report earnings from the third quarter of 2000 to the third quarter of 2001 that were almost $1 billion higher than should have been reported.

Many of the most significant transactions apparently were designed to accomplish favorable financial statement results, not to achieve bona fide economic objectives or to transfer risk.

Report of Investigation by the Special Investigative Committee of the Board of Directors of Enron Corp. (Powers Report), 1 February 2002, p. 4.

What could an analyst have seen beforehand? Not the restated figures. The 2000 10-K showed revenues up from $40,112 million to $100,789 million in a year, total assets nearly doubled from $33,381 million to $65,503 million and debt up from $8,152 million to $10,229 million. Rapid growth is not a red flag in itself; growth financed by debt, with earnings that depend on transactions with related parties, is the pattern the list below is built to catch.

Red flags

No single sign proves anything. Several together are a reason to read the notes closely and to ask management questions before relying on the profit.

• Receivables or contract assets growing faster than sales. Revenue is being recognised before customers pay, or customers are paying more slowly.
• Operating cash persistently below net income, year after year, without an investment or growth story that explains it.
• Costs capitalised that peers expense, such as start-up, advertising or routine maintenance costs, or useful lives longer than peers use. Capitalising moves the cash outflow into investing and the expense into later years.
• One-off gains inside operating income: sales of property, insurance settlements, the release of provisions.
• Related-party transactions and entities kept off the balance sheet, especially when the related party is an officer of the company.
• Changes in estimates or policies, such as longer useful lives, lower bad-debt rates or a new revenue policy, that raise profit in the year they are made.
• Dividends or buybacks paid while free cash flow is negative, financed by debt.
• Non-GAAP “adjusted” earnings that drift further above GAAP net income each year.

Lantern trips two of them in 2027: a contract asset that grew from nothing to $30,840, and a $2,000 dividend paid in a year of negative free cash flow. Neither is improper. Both belong in Coastline’s file. Analysts who want a more formal screen sometimes use statistical models such as Beneish’s M-score, which combines several of these signals into one score; like Altman’s Z-scores in the next lesson, it flags firms for review rather than proving anything.