We can never fully escape our childhood experiences, both good and bad. This old axiom receives dramatic new confirmation in research published in the November 2015 issue of the Economic Journal, which examines the long-term effects of periods of hunger during childhood in Germany after the Second World War.

The main finding of the study is that even 60 years after the hunger episode, those who suffered from hunger significantly raised their food consumption and had worse health compared with those who did not suffer from hunger. Those worse health outcomes included higher rates of obesity.

The researchers – Iris Kesternich, Bettina Siflinger, James Smith and Joachim Winter – conclude that adults'' dangerous over-consumption of food arose from fear of repeating the experience of hunger from their childhood years. The evidence shows that the worse health outcomes were concentrated in low-income households where the fear of repeating the experience of hunger may have been most realistic.

In the years immediately after the war, Germany was divided into four distinct occupation zones: American, British, French and Russian. Food was scarce due to the disruption of traditional food supplies, a large influx of refugees and deeply disrupted markets because there was no currency or trade between the zones. As a consequence, each zone rationed the amount of food that could be consumed – and rationing was particularly severe in the French and Russian zones.

The authors analyse data in which a random sample of people over age 50 living in Germany in 2008 were asked about their lifetime experiences. The questions included whether they had any experience of hunger in post-war Germany when they were children.

The authors then match those self-reports of childhood hunger with data on the actual monthly caloric rations that were imposed in the four administrative zones. They show that retrospective self-reports of childhood hunger closely match the severity of the caloric rations. When rations were more severe, more people reported they experienced hunger as a child.

The authors then find that the periods of severe childhood hunger had a very negative effect on the health of these people when they were much older. These negative outcomes included higher body weight and higher rates of obesity.

The reason for these negative health effects seems to be that those individuals who experienced hunger as children consumed much more food than would be normal for people with their adult incomes. Essentially the fear of repeating the experience of hunger from their childhood years induced them to over-consume food to the point where it was dangerous to their health.

These effects on over-consumption of food as an adult were concentrated in low-income households where the fear of experiencing hunger again may have been realistic. The effects on changing future food consumption behaviour were present when hunger was experienced at all childhood ages, and they were largest among those who experienced hunger in their teenage years.

In summary, adverse life events during childhood may affect not only adult health but also individuals'' behaviour over their whole lives. While health effects of shocks in utero and at various stages of childhood on adult health are well established, most of the existing research literature focuses on biological channels, with some recent studies also investigating the effects of shocks on cognitive skills.

The new research demonstrates that it is not only people''s biology that is changing: individual behaviour over their entire lives may be permanently altered.

''Individual Behaviour as a Pathway between Early-life Shocks and Adult Health: Evidence from Hunger Episodes in Post-war Germany'' by Iris Kesternich, Bettina Siflinger, James P Smith and Joachim K Winter is published in the November 2015 issue of the Economic Journal. Iris Kesternich is at the University of Leuven. Joachim K Winter is at the University of Munich. James P Smith is at the RAND Corporation. Bettina Siflinger is at the University of Mannheim.