The $1 Test

When a company earns a profit, its managers face the most consequential decision in capitalism: send the money to the owners, or keep it. Arche’s standard is explicit: keeping it is defensible only when there is a reasonable prospect, supported by historical evidence and thoughtful forward analysis, that every dollar retained will create at least one dollar of market value. Otherwise, return it.

Why this test is so clarifying

Retention is automatic at most companies — dividends get debated, but keeping the money never does. The $1 test drags that default into the light. A company that retained $50 billion over a decade while its market value rose $30 billion didn’t “invest in the future”; it destroyed $20 billion of its owners’ capital, one plausible-sounding project at a time.

Managers who deploy retained earnings on economically unattractive terms penalize owners twice — once through poor returns, and again because the market learns to discount every dollar the company touches.

How we compute it

For each company: the change in market capitalization over the measurement window, divided by cumulative retained earnings (net income minus dividends paid) across that window. A ratio of at least 1.0 passes; well-run compounders often show $3–6 of value per retained dollar. The window adapts to available price history and is always disclosed beside the number, because the test is meaningful over years, not quarters. Arche applies it over a disclosed multi-year window rather than treating a short-term price move as evidence.

Two honest caveats. The test borrows Mr. Market’s scoreboard, so at market extremes it can flatter or slander a decade of work; read it alongside the returns-on-capital record, not instead of it. And for REITs, which are required to distribute nearly everything, the test is waived — there’s almost nothing retained to judge.

The buyback corollary

Retention’s sibling is the share repurchase, and it has its own price test: repurchases make sense only when the stock trades at a material discount to a conservative estimate of intrinsic business value. A buyback below value is retention that passes the $1 test instantly; a buyback above value is a quiet transfer from continuing owners to sellers. Our Management pillar scores both: the trajectory of the share count, and — in the research reports — whether the buying happened when the price was low or merely when the cash was plentiful.