Music helps us estimate how healthy a country’s economy is, a new study surprisingly reveals. More to the point, national unemployment rates can predict the negative emotional content in lyrics in a country’s songs.

Songs have a very powerful emotional component, and past research has shown that people tend to listen to tunes that match their current moods or preoccupations. Starting from that chain of thought, a new paper aimed to find if music can be used to estimate the socioeconomic health of a community (in this case, a country).

The team worked with popular song lyrics from the US and Germany and report that unemployment rates predicted feelings of anger portrayed in songs in both countries.

Rage against the economy

“This study aimed to examine how sentiments in top songs coincide with changes in national unemployment rate. In particular, we focused on three common negative emotions (i.e., anxiety, sadness, and anger) expressed in lyrics,” the researchers say.

For the study, the team used a text analysis program called the Linguistic Inquiry and Word Count (LIWC) to trawl through the lyrics of the top 10 most popular songs in the US and Germany between 1980 and 2017. Songs with no lyrics were removed from the analysis.

The final sample of US songs included 370 lyrics (149,660 words) while the German sample totaled 366 lyrics (120,076 words). From there, the team used the LIWC to estimate emotional content in three categories: anger, sadness, and anxiety. They also looked at how word frequency denoting each emotion compared to unemployment rates in each country at the time these songs were written.

At first, the team writes, there seems to be no discernible link between unemployment rates and such negative emotional content in music. However, after controlling for other. elements that impact individuals’ economic prospects, especially ones that tie into inflation (GDP per capita, housing prices, inflation, and population density), the unemployment rate showed itself to be a “significant predictor” of anger content in US lyrics, they report.

German lyrics fared similarly: there was no immediate discernable link between unemployment and sadness or anger. After controlling for the same indicators, however, it was a good predictor of anger in lyrics.

As to why this dynamic forms, the authors have two theories. The first one is that socioeconomic factors can impact the emotional state, and thus behavior, of consumers. High rates of unemployment can nurture feelings of stress and anger, and consumers might favor songs that reflect such a state, driving them up in the charts. The second theory is that such factors impact artists and composers who transpose their feelings of stress and anger into their work.

The explanation could, of course, lie somewhere in the middle of these tho theories.

One interesting tidbit of these findings is that sadness or anxiety didn’t seem to change in response to employment rate — suggesting that anger is the primary public response to poor economic prospects.

“This is consistent with preliminary research illustrating that unemployment can lead to various affective responses, but the central emotional response is anger when the adversity is attributed to external causes,” the paper reads.

One of the study’s most obvious limitations is that it only looked at the lyrical component of songs, ignoring the musical frame around them. This frame could alter the emotional message being conveyed by the songs. It also focused on two countries in the Western world. Thus, it is unclear whether dramatically different cultures would show the same response.

In the future, the authors plan to control for “melodic attributes” in songs as well, in order to better gauge their emotional content.

The paper “Unemployment Rate Predicts Anger in Popular Music Lyrics: Evidence From Top 10 Songs in the United States and Germany From 1980 to 2017” has been published in the journal Psychology of Popular Media.