Most of us take a short break for coffee, a longer break for lunch, a longer break from late afternoon to the following morning, a longer break at weekends, a longer break for holidays, and then a very long break indeed when we get old.
There are two questions:
1. Why do we bunch our leisure?
2. Why do we take a great big, and increasingly big, bunch of leisure at the end of our lives? (We call it "retirement".)
People are supposed to be consumption-smoothers. There is supposed to be diminishing marginal utility of consumption, so you maximise your lifetime utility if you smooth your consumption over time. If we had high consumption with low marginal utility in some periods, and low consumption with high marginal utility in other periods, we could increase our total lifetime utility by reducing consumption when marginal utility is low and increasing consumption when marginal utility is high.
Why doesn't this same argument apply to the consumption of leisure?
If real interest rates were higher than your subjective rate of time preference, then you would choose a consumption path that is smoothly growing over time. You would consume more when old than when you were young. For all other consumption goods, some people act a bit like that, and others don't. Some people consume a bit more when old than when young, and others consume a bit less when old than when young. But when it comes to the consumption of leisure, we don't act at all like that. We consume roughly the same amount of leisure every year. And then we suddenly go on a massive consumption binge for the rest of our lives. Why?
As we have gotten richer over the last two centuries, we have chosen to spend a greater proportion of our lives consuming leisure rather than working. But I think I'm correct in saying that by far the biggest increase in our leisure has been in that big bunch of leisure at the end of our lives. Most people used to work until they died. Now most people stop working and then consume a decade or two of leisure before they die. Proportionately, the growth in leisure taken in retirement massively exceeds the growth of coffee breaks, lunch breaks, evenings, weekends, and holidays. The income elasticity of demand for retirement leisure massively exceeds that of all other forms of leisure. Why?
Non-convex (non-concave? whatever) production technology must be part of the story. There are fixed costs of commuting to work, so we work in bunches.
Non-convexities in the consumption of leisure must be another part of the story. There are fixed costs of commuting to good places to go canoeing, so we take leisure in bunches.
But why should these non-convexities have such a massively bigger effect for retirement? How do they explain the very big income elasticity of demand for that great big bunch of leisure we call "retirement"?
The obvious answer is that our productivity falls as we age. So it makes sense to consume more leisure when the opportunity cost is least. But there's also an obvious problem with that explanation. If our productivity at work falls as we age, maybe our productivity at leisure falls as we age too. Which will get worse more quickly: my ability to give an economics lecture; or my ability to portage a canoe?
Obsolescence of human capital?
Just some random thoughts. This is not my area at all. Obviously inspired by current talk of raising the age for eligibility for OAS/GIS. But also inspired by my thoughts for my own future.