The value effect—the pattern of value stocks outperforming growth stocks—has been around for a lot longer than most people realize, contends an article in MarketWatch, and there is historical data to prove it. Researchers from Robeco Quantitative Investments in the Netherlands stretched all the way back to 1866 to create a database of individual stock prices, and were able to measure how value stocks performed compared to growth stocks that far back for the first time.

Previously, researchers had only studied as far back as 1926, most notably the Fama & French analyses from 1991. Those studies found that in the period from 1963-1991, value significantly outperformed growth. But from 1926-1963, the value effect was much weaker. The new research finds that the value effect was decidedly strong from 1866-1926, much more so than in the 1926-1963 era and after 1991.

However, the Dutch researchers relied on the dividend yield for their study, as the data on book values of publicly-traded companies just doesn’t exist going back as far as 1866. Fama & French used price-to-book ratio for their analyses, and that has since become the standard metric for measuring value against growth. But the Dutch researchers wrote that “though the lists of value and growth stocks constructed with the dividend yield are not identical to those constructed using the price-to-book ratio, there is a high degree of overlap.”

The new research indicates that the value effect stands on a more solid historical foundation than most realized, the article maintains. And even if value trails behind growth for a bit, it’s safe to predict that the value effect will likely reassert itself once again.