The highest-ranked companies of 1997

What this ranking cannot tell you

These are the companies that ranked highest in 1997 and are still listed today.

Companies that were acquired, taken private, or failed between 1997 and now were removed from our data and are missing from this page. That is not a rounding error and it points one way: it makes every historical ranking look better than the real one did, because the companies that did not make it are the ones that dropped out. We began preserving delisted companies in 2026; the ones lost before that are gone for good, so this limit is permanent rather than something we expect to fix.

The numbers are what we know now, not what was printed then.

Our data provider supplies each company’s current, restated financials. We cut them off at the date you are viewing, but we cannot un-know a later correction, reclassification, or an acquisition folded back into earlier years. So this is the ranking for that year computed from what we know today about the years up to it, which is not quite the ranking that existed at the time.

A year’s results only count once they were actually published.

Annual reports appear months after the fiscal year they cover, so we hold each year back by 90 days before letting it affect a ranking. Otherwise a company whose fiscal year ended days earlier would be scored on figures nobody could read yet. Ninety days is measured from real filings, but from recent ones, and companies filed more slowly further back — so for the oldest years here the allowance is, if anything, slightly short.

Ranking highly in a given year did not predict what came next.

We tested this directly, and published the results on our methodology page. Taking the fifty highest-ranked companies at a past date and holding them to today underperformed simply buying the S&P 500 at nearly every start date and holding period we have tried — and in some cases underperformed holding every company in our data equally. Where it did come out ahead, at the 2006 start over ten and twenty years, the margin was small enough to sit inside the difference between measuring on exact dates and measuring on our July-1 grid, so we do not read it as a win. Strong fundamentals tend to coincide with strong returns while they last, not to forecast the returns that follow. Read this as a record of what these businesses had done by then, never as a screen for what to buy.