If Census were to exclude Social Security benefits from income, the poverty rate for American seniors would jump from 14.6% to a whopping 52.6%—roughly 23.4 million people.
Consider three counterfactuals.
1. The government makes a sudden, surprising decision to renege on promised SS benefits.
2. The government tells all young people that SS is abolished for them, and they are on their own in terms of planning for their retirement.
The analysis of the Census data tells you what is probable under (1), but that is not a terribly interesting scenario for policy purposes. The interesting policy scenario is (2). In that case, when the hypothetical young people retire, their poverty rate could be either higher or lower than under SS, depending on how much they save and how well they invest that money.
It would be much higher for their attempts to save would be doomed to failure by reducing demand but could not increase investments with rates already at/near zero but only trigger deflation and increased unemployment. Societies as a whole cannot save.
“How they invest their money” will depend on what there is for “money” to do.
Look at the cash on the books of large production and distribution businesses whose managers simply will not redeploy the surpluses.
What’s the third counterfactual?