Third Way produced a chart here that I can only assume was designed to illustrate the point that it’s easy to mislead people if you represent government spending trends over a long time horizon in terms of dollars rather than shares of GDP:
But always a little bit taken aback by the notion of doing 70-year projections of things. 70 years ago it was 1940. Still ahead was Pearl Harbor, World War II, the Cold War, Medicare, civil rights, mass immigration from Mexico, computers, the fall of the Soviet Union, the Internet, etc.
The upshot is that I think it’s best to think about this sort of thing in very abstract terms. Currently, the elderly share of the population is expected to rise over time. Consequently, either future elderly individuals need to consumer less relative to the median individual than is currently the case or else taxes on future non-elderly individuals need to be higher or else spending on the non-elderly needs to be lower. Of course, those measures can also operate in various combinations. What’s more, different kinds of policy interventions can alter the demographic projections. If you have some kind of view as to what you like out of those abstract ideas, then you can get down to tinkering with exact tax rates and benefit formulae but thinking that you’re going to precisely nail decades worth of economic forecasts is bit of a mug’s game.