I was cleaning up my office yesterday when I stumbled across a little book called Why Yesterday Tells of Tomorrow, by Helmut Gaus.  A friend gave it to me years ago and I smiled when I saw the cover again.  I remembered it showed some of the most creative (some would say outrageous) data correlations I had ever seen.  One of them for example compared the average length of skirts with the number of suicides in France.  The point Gaus was making however was very interesting and it couldn’t be more actual today.  Are changes in the economy caused by changes in the mental state of society?

Gaus’ theory starts with the work of Russian statistician Nikolaj Kondratieff.  In 1925 Kondratieff discovered long macro economic waves that would repeat themselves every 50 years throughout history through alternating periods of growth and decline.  Click on the picture below for more details on the Kondratieff waves.

Kondratieff himself had to pay a heavy price for this discovery.  His predictions that the economy would eventually decline were not popular with Stalin who sent him to Siberia to be executed. 

Much has been written about whether the Kondratieff waves actually exist.  Not everyone is convinced but the truth is they have been reasonably predictive, to this day.  Apparently we have just entered a Kondratieff Winter – a period severe of depression that can last up to 20 years.  Let’s hope these waves are not that predictive!

Gaus was most interested in the causes of the waves which most thought to be macro economic.  Gaus was convinced that the real drivers were psychological.  He argued that the Kondratieff waves coincided and in some cases even followed what he called anxiety waves which are predictive of changes in tastes, values and behavior. 

To prove his point he gathered data on women’s fashion trends.  He found that in periods of high anxiety women wear less color, fewer patterns, lower necklines and longer skirts.  In periods of lower anxiety they wear brighter colors and patterns, higher necklines and short skirts.  This allowed him to create an anxiety wave based on the prevalence of these fashion trends over time. 

He found startling correlations between anxiety levels and all sorts of indicators such as marriages, births, employment levels, suicides and investment levels over time.  He also found striking correlations between fashion indicators and the Kondratieff waves.  The graph below for example shows the average width of the waist and the Kondratieff wave for the US from 1791-1936. 

The idea that fluctuations in the economy are caused by the collective levels of anxiety is interesting but hard to prove.  Data on the mental state of society is scarce, which is why Gaus used data on fashion as a proxy.  There is however a relatively new data source that holds exteremely rich informatoion on what’s on people’s minds – it’s the data held by search engines.  Knowing what people search on and how that changes over time could potentially lead to a barometer of society’s mental state.  Maybe the folks at Google would be interested in doing some research in this area.  Saying that though, they’re probably doing this already … .