Working capital and the current ratio, explained

August 29, 2026 · 9 min readfundamentalseducation

Working capital is one subtraction: current assets minus current liabilities. The current ratio is those two numbers divided instead of subtracted. Both ask the same thing. Does what turns into cash over the next twelve months cover what comes due over the next twelve months?

Take a real answer. Target's second-quarter report puts total current assets at $20.928 billion against total current liabilities of $21.180 billion as of 1 August 2026. Working capital: negative $252 million. Current ratio: 0.988. The same release reports $1.877 billion of net earnings for the quarter on $26.539 billion of sales. So the balance sheet says the next year's bills aren't covered by the next year's assets, and the income statement says the business just cleared close to two billion dollars in thirteen weeks. Both readings are correct, and a rule that reads only one of them will pick the wrong companies.

This is the fourth piece in a series building up the vocabulary our theses use. The third one walked a balance sheet end to end. This one zooms into the top of it, where the twelve-month claims live.

What counts as "current"

Current means within twelve months, or within one operating cycle when a business takes longer than a year to turn cash into goods and goods back into cash.

Current assets, roughly in order of how fast they convert:

cash and cash equivalents
short-term investments
receivables: billed and awaiting collection
inventory: goods awaiting a sale
prepaid expenses

Current liabilities:

accounts payable: supplier invoices
accrued expenses: wages, taxes and interest already incurred
deferred revenue: cash collected for goods you still owe
short-term borrowings and the current portion of long-term debt: the slice of the debt stack due inside the year

That split isn't universal, and the SEC's rulebook admits it in the caption itself. Regulation S-X lists the balance sheet captions a company must present, and the one for total current assets reads "Total current assets, when appropriate." Total current liabilities carries the same qualifier. Hold onto those two words; we come back to them.

The arithmetic, on a filing you can open

Two lines off the statement, two operations:

working capital = current assets - current liabilities
current ratio   = current assets / current liabilities

For Target on 1 August 2026 that's 20,928 minus 21,180, a deficit of $252 million, and 20,928 divided by 21,180, a ratio of 0.988. The same release prints the year-ago column: $19.034 billion of current assets against $19.223 billion of current liabilities, a deficit of $189 million and a ratio of 0.990. Two years running, a small deficit and a ratio that barely moves.

Now open the composition, because the totals hide it. Of that $20.928 billion of current assets, $13.249 billion is inventory: 63 percent of the pile. Cash is $5.411 billion. On the other side of the ledger, accounts payable is $13.306 billion, $57 million more than the entire inventory balance sitting in the stores and warehouses. Target's suppliers are funding the whole shelf.

That's the negative-working-capital model, and it's a feature of the business. Merchandise arrives on credit, customers pay at the register within days, and the supplier invoice comes due weeks later. The float between those two events is funding the company gets for free, and it shows up on the balance sheet as a current ratio parked below 1 forever. One condition is attached: it works while the goods keep moving. A retailer whose inventory stops selling still owes the payables on the original schedule.

When 0.99 is a warning

The ratio can't see any of what we just walked through. It counts dollars in the numerator without asking how quickly they become cash, and dollars in the denominator without asking when they come due.

Pull the inventory out of Target's numerator and the ratio falls to 0.36 ($7.679 billion over $21.180 billion). That's the quick ratio, the same test with the slowest current asset removed. For a retailer that turns its shelves several times a year, the fuller number is the fair one. For a company sitting on goods nobody has ordered, 0.36 is closer to the truth. Identical formula, opposite conclusions, and the balance sheet by itself won't tell you which company you're looking at. Inventory and sales trends will.

The denominator hides things too. Short-term borrowings and the current portion of long-term debt come due on a calendar date whether or not the receivables collect and the inventory sells. That's the gap between profit and solvency: profit is what a period produced, and solvency is whether the money is in the account on the morning a payment clears. A company can report earnings for years and still fail to make one payment on a Tuesday, at which point the earnings history stops mattering.

Working capital is also the bridge between those two worlds. An inventory build or a stretched receivable consumes cash that the income statement never mentions, which is why the cash flow statement adjusts for changes in working capital on its way to operating cash flow. When you see profits and cash generation drifting apart, the working capital lines are usually where the drift is happening.

Where "current" stops meaning anything

Back to "when appropriate." Plenty of balance sheets never split current from non-current at all, for two reasons worth knowing.

The first is a long operating cycle. A homebuilder buys land, entitles it, builds on it and closes the sale over a stretch that runs well past twelve months, and its inventory is dirt and half-finished houses. Builders generally present the statement unclassified, with no current subtotals anywhere on it, because there's no defensible place to draw the twelve-month line.

The second is banking. Deposits are payable on demand while loans mature over decades, so sorting a bank's balance sheet into current and non-current buckets would make every bank in the country look insolvent every day of its life. Regulation S-X puts banks and insurers under separate articles with their own captions.

Which is why both bundled templates that touch the current ratio exclude those sectors up front. The Piotroski F-Score template's universe block carries the comment "Bank/REIT leverage and 'current' ratios don't mean what the signals assume," and Graham Net-Net says the same about financial balance sheets making current assets meaningless under its lens. Where a filing has no current-asset line at all, the feature returns null and the quality guard flags the name out of the run. A missing subtotal should cost a company its score, because the alternative is inventing coverage the filing never claimed.

How the bundled theses read it

Two of the five shipped templates use the current ratio, and they use it in opposite directions.

Graham Net-Net scores the level. Its solvency criterion wants a current ratio of 2.0 or better for full marks, 1.2 for partial credit, on the logic that a net-net that can't pay its near-term bills gets liquidated at fire-sale prices and the floor you thought you bought doesn't hold. The feature also handles the odd filer with no current liabilities at all:

cr: if(cl > 0, ca / cl, if(ca > 0, inf, null))

Zero current liabilities with assets on hand is infinite coverage. Zero of both is missing data, and those deserve different answers. Worth remembering that the net-net calculation itself goes further than the current ratio does: it nets current assets against total liabilities, so long-term debt comes out too.

The Piotroski F-Score template scores the change. One of its eight signals asks whether liquidity improved against the same fiscal quarter a year earlier (why the same quarter and not the adjacent one):

cr:       if(newest(current_liabilities) > 0, newest(current_assets) / newest(current_liabilities), null)
cr_prior: if(lag(current_liabilities, 4) > 0, lag(current_assets, 4) / lag(current_liabilities, 4), null)
f_liq:    cr > cr_prior

Put Target's two columns through that and the signal comes out false: 0.990 a year ago, 0.988 now. No point, on a difference of two thousandths. Binary signals are crude on purpose, trading magnitude for robustness on noisy quarterly data, which is why the score adds up eight of them and the bundled template gates at 7 of 8 before a name survives. One signal deciding anything by itself would be an accident.

Writing your own liquidity rule

Once you can read the two lines, the rule is short. The shape below is illustrative and the bundled templates are the reference for exact fields:

# "Can it pay next year's bills?" - illustrative
params:
  cr_floor:    1.2           # coverage that counts as adequate
  quick_floor: 0.8           # coverage without selling any inventory

features:
  ca:       newest(current_assets)
  cl:       newest(current_liabilities)
  inv:      newest(inventory)
  wc:       ca - cl
  cr:       if(cl > 0, ca / cl, null)
  quick:    if(cl > 0, (ca - inv) / cl, null)
  cr_prior: if(lag(current_liabilities, 4) > 0, lag(current_assets, 4) / lag(current_liabilities, 4), null)

criteria:
  liquidity:
    rules:
      - { when: "is_null(cr)", score: 0, flag: no_current_split }
      - { when: "cr >= $cr_floor and quick >= $quick_floor", score: 2 }
      - { when: "cr >= $cr_floor", score: 1 }
      - { else: 0 }
  improving:
    rules:
      - { when: "is_null(cr_prior)", score: 0, flag: no_year_ago_filing }
      - { when: "cr > cr_prior", score: 1 }
      - { else: 0 }

gate:
  - cr > 0                   # no current split, no score

Both criteria open with the null case: a missing line scores zero and raises a flag, never a default. And the interesting experiment isn't the thresholds, it's the sign of the rule. A version that rewards a high current ratio and a version that treats a low one as evidence of supplier financing are different theses about what liquidity means, and they'll surface different companies. You can write both and test both, which beats arguing about it. The DSL reference has the full field vocabulary.

What the number buys you

Working capital and the current ratio are a snapshot of one date: the claims arriving inside a year, and the assets meant to meet them, before a single item has converted. They tell you how much slack a company has and who is financing whose inventory. They can't tell you whether the goods sell, whether the customers pay, or whether the lender renews the credit line next quarter, and no amount of decimal places will change that.

What they do have going for them is that they're two lines and one operation, checkable against the filing in under a minute. That's a low bar and a valuable one. Build the rule, see which companies it catches, then go read their filings and find out whether the ratio was telling you anything.

Next in the series: gross debt, net debt, and how leverage reads on a screen.

Research tooling, not investment advice. Nothing here is a recommendation to buy, sell, or hold any security. Screens, scores, and backtests are informational only; backtested results are hypothetical, exclude costs such as commissions and slippage, and do not guarantee future results. Verify against primary filings and make your own decisions.

Want to try this on your own rules? Quantery is free for 14 days: the full app, no card required.

← All articles