Can you remember what you ate yesterday? If asked, most people will be able to dredge up a vague description of their main meals: breakfast, lunch, dinner. But can you be sure you’ve noted every snack bar en route to the car, or every handful of nuts at your desk? Most people will have the lingering feeling that they’ve missed something out.
We originally had this suspicion back in 2016, puzzled by the fact that national statistics showed calorie consumption falling dramatically over past decades. We found reliable evidence that people were drastically under-reporting what they ate – and the problem was getting worse over time.
Now the Office for National Statistics has responded to our report by confirming its findings: we are consuming 50% more calories than our national statistics claim.
Why is this happening? We can point to at least three potential causes. One is the rise in obesity levels itself. There is good evidence that under-reporting rates are much higher for obese people. The main reason for this seems to be that obese people simply consume more food, and thus have more to remember.
Another cause is that the proportion of people who are trying to lose weight has been increasing over time (from 40% in 1997 to 50% in 2013, for example). People who want to lose weight are around 10 percentage points more likely to under-report their eating – regardless of whether they are overweight or not. This may be driven partly by self-deception or “wishful thinking”.
The final potential cause is an increase in snacking and eating out over recent decades – both in terms of how often they happen and how much they contribute to our overall energy intake. The amount of time spent eating out doubled between 1975 and 2000, for example. By 2015, one in five meals was eaten outside the home. This trend means it is more difficult for us to keep track of what we eat, not least because we have to remember more eating events. Again, there is evidence for this – food consumed out of the home is one of the most poorly recorded categories in surveys.
We will be presenting these findings in a BBC documentary called The Truth About Obesity. However, we want to stress that they are not just interesting statistical nuggets. First, the differences we are talking about are huge – men are consuming 1,000 more calories a day than previous figures suggested. This is the equivalent of not reporting that you ate a whole pepperoni pizza.
Second, these statistics guide and underpin policy. There is much concern about the health effects of obesity, but what do we do to reduce it? Based on the old numbers, our food consumption would not be seen as the problem: it seems to be below recommended intakes, and falling. So policymakers may be tempted to look elsewhere – perhaps to increasing physical activity instead. We showed that this is not the most effective way of preventing obesity, given that it takes much more effort to burn calories than consume them.
So, what’s the takeaway? For statistics, we should invest in more accurate measurement options – these do exist, but they can be expensive. For policy, we need to focus on options that make it easy for people to eat fewer calories. If people do not know how much they are eating, it can be really hard for people to stick to a diet. Instead, we should be looking for new ways to incentivise and achieve food reformulation. If this works, then people would not need to try to eat less – it’s just that what they eat wouldn’t have the same impact on their waistlines. And it won’t matter so much if they can’t remember whether it was a muffin or a croissant yesterday morning.
• Michael Hallsworth is director of health and tax at the social purpose company, Behavioural Insights Team
