Stable Cultural Values and NormsOnce upon a time—a project examining links between fairy tales and economic growthDoing the Research series

27 August 2025, by Newsroom editorial office

Photo: iStock/ZU_09

Values and norms have been passed on in legends and fairy tales for centuries. A research team led by Prof. Dr. Stefan Voigt from the Faculty of Law used artificial intelligence to investigate if this is still a growth factor in today’s economies.

At first glance, fairy tales and law or economics seem to bear no relation. So what inspired this research topic?

I wanted to know how to establish the long-term relevant norms and values of a society that are decisive for the growth of an economy. For centuries, fairy tales were mostly passed down orally and only the most successful ones are available in writing today— successful, meaning they have been used to reflect prevailing beliefs and to communicate them to children.

I started thinking about exploring this aspect of fairy tales over 30 years ago, when everything revolved about the growth prospects of Eastern European countries after 1990. At that time, however, it was impossible to read and prepare thousands of fairy tales for a code comparison. The emergence of large language models, such as ChatGPT, created new possibilities.

Your research is based on 27,000 legends and fairy tales— what were the criteria for texts and where did you find the fairy tales?

Our study was aimed at recording values and norms that were established a long time ago. We thus compiled fairy tales that had been recorded in written form by the end of the 19th century. The records come from folklorists such as the Brothers Grimm as well as from colonial officers or missionaries who recorded fairy tales in the colonies. We compiled the various sources and—if not yet done—digitized and translated them into English.

You examined the texts for 7 groups of norms and values: Cooperation, selflessness, work ethic, entrepreneurship, tolerance of dissent, rejection of discrimination, and questioning authority. Can you give an example of how you analyzed the fairy tales in terms of these categories?

We think that these individual factors have a positive impact on economic development. Each factor was recorded by so-called variables representing the characteristics of that norm. The work ethic, for example, draws heavily on Max Weber’s Protestant thesis, so we examined the fairy tales for hard work, attention to detail, honesty, patience, keeping promises, and punctuality.

However, instead of checking whether specific keywords were mentioned, we wrote a prompt for each variable so ChatGPT had 4 options to evaluate the story. For the abovementioned example these are: Honesty is encouraged, honesty is not encouraged, honesty does not occur, or ChatGPT is not sure.

In total, we had 77 variables and applied a corresponding number of prompts to each fairy tale. Thirty-six were included in the 7 main categories of our current evaluation. We now know whether honesty plays a role in any of the 27,000 fairy tales and if so, whether it is encouraged or not.

How else did you use the data?

Of course, not all aspects are covered by every fairy tale. Cooperation, for example, is more often a subject than rejection of discrimination. However, based on the data points you can create an average for each variable. If you organize the fairy tales by countries, you can see whether honesty is promoted as a value or rather negatively connoted in comparison with the average.

Our work focused on the question of whether past presentations of values are correlating with differences in economic growth today. How did you proceed?

We included our values into a statistical calculation of economic growth. These models traditionally include factors such as work, investment, and level of education, and also geographical country parameters. These figures are available for the 140 countries we analyzed.

For each of the 7 main categories, we have calculated a value that indicates, for example, to what extent the willingness to cooperate in the fairy tales of a country deviates from the average of all fairy tale. Positive values translate into an above-average willingness to cooperate, whereas negative values indicate a below-average willingness to cooperate.

By adding these different variables to an econometric model, we can establish whether individual variables contribute significantly to explaining the different growth rates, and if so, to what degree exactly.

What did you find out?

We looked at the period between 1990 and 2019, as we have comprehensive data for many countries from that period. Our research shows that the examined differences in the manifestations of our 7 norms and values categories correlate significantly with the varying growth rates. Countries where the willingness to cooperate was strengthened in fairy tales at the turn of the 19th century are still growing stronger to date. Where this value did not gain a positive connotation, growth is lower today.

You will be sharing this Historical Values Survey dataset with other researchers. What other questions could be answered based on your research findings?

It is the first dataset of historically widespread values and norms based on 27,000 fairy tales from 172 countries that are home to 99 percent of all people. This global dataset could be of interest to a range of disciplines, not only for nation-state research, but also for regional investigations.

In the next step we could assess to what extent the historical values examined have changed to date. For this we could compare our data with the World Values Survey that has been conducted regularly since 1981 to examine the perception of values and norms around the globe.