Good company reputation is valuable for shareholders
Maine-survey ranks Finnish listed companies yearly by their reputation among investors. To find out how competent reputation management affects shareholder value, we backtested how a portfolio of top-ranked companies performs compared to other listed companies.
Arvopaperi, a Finnish investing magazine, and Burson-Marsteller, a public relations agency, have conducted a reputation survey yearly since 2001. The survey asks private investors to rate publicly traded companies by several criteria like corporate culture, products and services, management and ethics. To compare the best-reputed companies against the other ones, we simulated two investement strategies.
- “Top Ten” contained the 10 top-ranked stocks from five consecutive surveys: Maine 2010 – Maine 2014. Stocks were bought at the closing price in the same day when the Maine survey was published, held for a year and sold when the next year’s Maine survey became public.
- “Everything Else” had all other large-cap and mid-cap stocks, which were listed in OMX Helsinki exchange 1.1.2016.
“Top Ten” didn’t have small caps so we left them out from “Everything Else” also to avoid testing the small firm effect. Both portfolios were equal-weighted at buying price, so we didn’t do the apples-to-oranges comparison to market capitalization weighted indices. The comparison takes into account dividends, spin-offs, splits, reverse splits and buyouts, but doesn’t take into account survivorship bias: stocks, which were delisted before 1.1.2016 are ignored (except for Vacon) because our data source didn’t have them.
|Period||Top Ten||Everything Else|
|Maine 2010: 28.10.2010 – 27.10.2011||-4.6%||-15.8%|
|Maine 2011: 27.10.2011 – 25.10.2012||+4.7%||-1.9%|
|Maine 2012: 25.10.2012 – 24.10.2013||+25.9%||+29.2%|
|Maine 2013: 24.10.2013 – 21.11.2014||+4.0%||+13.5%|
|Maine 2014: 21.11.2014 – 19.11.2015||+30.2%||+19.9%|
|28.10.2010 – 19.11.2015||+70.1%||+45.1%|
Between the publication of Maine 2010 and Maine 2015, the “Top Ten” would have yielded 70%, while “Everything Else” would have returned 45%. Annualized, “Top Ten” yielded 3 percentage points more, but the data set is too small and the yearly variation too big for this figure to be meaningful. The result is consistent with scientific research about the ability of crowdsourced investor reputation to predict stock-market returns. The paper Wisdom of Crowds: The Value of Stock Opinions Transmitted Through Social Media looked at the Seeking Alpha website, where investors write analysis about listed companies. They used automatic sentiment analysis to find out how positively or negatively analysts and commenters wrote about companies. Positive comments, indicating good reputation among investors, predicted both higher stock-market returns and positive earnings surprises.