In August 2013, I wrote a column called “Honesty Did Pay.” Based on anecdotal evidence, I thought there might be a positive correlation between a country’s ethical norms and its stock market performance. The connection struck me as logical. Presumably, companies that retained their profits rather than paying them as bribes would post higher earnings. In addition, those managed by honest executives would probably take better care of outside shareholders.

The article compared the trailing 20-year equity returns for 21 countries against each nation’s perceived level of honesty, as measured by Transparency International’s “Corruption Perceptions Index.” The results supported my hypothesis. Summarizing the findings, I wrote,

“Transparency International’s rankings have lined up quite nicely with the past 20 years of stock performance. They almost certainly won’t line up as neatly over the next 20 years, but perhaps there will be enough correlation to make the rankings useful.”

Round 2: The next decade

Now that we’re halfway to the 20-year mark, let’s see whether that prospect has occurred. (My patience, it turns out, lasts only for one decade.) For this test, I expanded the candidate list, courtesy of the growth in international indexing. Morningstar publishes 10-year track records for 43 countries. I used them all.

I ranked each nation’s annualized total returns, computed in U.S. dollar terms, from 1 to 43. The country with the highest performance—Denmark—received a grade of 1, while that with the lowest total (Colombia) was ranked 43. I repeated the process using 2012 Corruption Perceptions Index ratings. Denmark also scored first on that account, with Indonesia at 43.

The following chart plots each country’s rankings on those two measures. When moving along the y-axis, dots that trend upward indicate a positive correlation, meaning that higher stock prices were associated with greater honesty. Dots that trend downward suggest a negative correlation. Finally, dots that show no visible pattern indicate that that correlation was too slight to warrant attention.

stock market performance

The picture is clear: Honesty pays, once again. Almost certainly, that outcome was not accidental. The p-value for the regression was 0.3%, indicating that such a strong positive correlation would randomly occur only three times in every 1,000 draws. (The academic standard for a “statistically significant” p-value is far more lenient, at 5.0%.)

The local-market view

It’s possible that these results owe to currency movements rather than to stock behavior. Perhaps the higher-returning countries had powerful currencies, while the laggards were sunk by declining exchange rates. Whether the answer to that question should affect the conclusion is unclear—does it matter whether honesty is associated with rising currencies or rising local equity prices?—but the matter nevertheless deserved investigation.

I reran the numbers, in this instance measuring equity performance in local currencies as opposed to U.S. dollars. (For this calculation, I omitted Turkey, Egypt, and Brazil, as their high local-currency returns were consumed by equally high inflation rates.) The outcome is displayed below.

Local currencies

No meaningful change. The dots still rise, and the p-value for the regression, although higher than before, remains statistically significant. As it turns out, the positive correlation between stock performance and corruption level was not primarily caused by currency movements but instead by stock price behavior.

Developed versus emerging markets

My inquiry was not yet complete. Whereas the 2013 study included few emerging markets because data was not available, the current set contained many such countries. This led to a further question. With rare exception—Singapore being the most notable example—emerging markets consistently rank worse on the Corruption Perceptions Index than do developed countries. Was the distinction between emerging and developed markets the true underlying factor?

Easy enough to test. I sorted the 43 countries into those two buckets, based on classifications from the United Nations, then ran the numbers for the developed nations in U.S. dollar terms. The positive correlation did not entirely disappear, but the tendency was distinctly faint. More to the point, the p-value soared to 32.2%.

developed markets

This time, history did not repeat. Whereas in 2013 countries placing higher on the Corruption Perceptions Index tended to outgain the developed-nations average, the 2023 totals exhibit no such tendency. It was good to be a developed market, but after that, there is little to be said.

The chart for the emerging markets, which I will not show, amounts to much the same. Nobody this side of a politician would find its very modest positive correlation worth touting. The bottom line is that, when sorted by developmental status, the relationship between low corruption and high stock returns vanishes.

Conclusion

When I created this column’s first chart, I was surprised that the pattern that I had observed from 1993 through 2013 had persisted so powerfully through the following decade. Investment history is rarely so reliable. As it turned out, the continuity was largely an illusion. During the first period, differences in ethical practices within the developed markets mattered. During the second, they did not. Nor did differences within the emerging-markets nations.

While the honesty hypothesis was no longer useful for selecting individual countries, it remained applicable when viewed more broadly. My 2013 article concluded:

"International stock investors typically pursue economic growth, not cultural honesty. After all, the ETFs of the BRIC countries of Brazil, Russia, India, and China didn’t become far, far more popular than the ETFs of Northern Europe because the former countries have cleaner governments. Investors may wish to adjust their priorities."


Indeed, they may.