4 ms·
Look at the graph of spending on children, look at how the lines for the bottom 3 quintiles have gone down since the '80s and '90s. That's very disheartening to
by InclinedPlane 8y ago
Look at the graph of spending on children, look at how the lines for the bottom 3 quintiles have gone down since the '80s and '90s. That's very disheartening to see, parents are going to try their hardest to give their kids the best chances, and yet 60% of families have been squeezed by the economy into providing less for their children than previous generations. This despite inflation adjusted per capita GDP nearly doubling since the early '80s. The economy is growing, wealth is being created, but it's being funneled into a tiny portion of the population. For so long "the rich get richer while the poor get poorer" has been a common refrain without being strictly true, but now it's become true.
- souprock 8y agoI don't know that the graph tells this story. It could be that the very slight decline in spending for the lower quintiles is from getting help that isn't counted (free preschool for example) or from products being more affordable (measuring money is not equivalent to measuring the value gotten) or from more people being able to avoid childcare expenses by being home. It could be that the major increase in spending for the top quintile is wasted. Maybe the kid wants to own a horse. All sorts electronic things eat up money while delivering little value.
- InclinedPlane 8y agoYeah, you're probably right, I'm sure it's just horse ownership and free preschool rather than income stagnation, exploding debt in the under 35 age bracket (much of it due to student loan burden), and massively increasing housing costs.
- thisisweirdok 8y agoFree preschool isn't incredibly widespread, but I wouldn't be surprised if increasing daycare costs are causing more people to drop out of the workforce to care for their kids. It's certainly complex.
- jacobolus 8y agoMy suspicion is that a big part of this is declining family size among all income brackets. The reported numbers are per household rather than per child (excluding households with no children). Amount spent per child might be flat or even still even be slightly rising for these groups (though obviously nothing like the booming spending of the wealthy).