Right, Kevin, that is “something close to the definition of social engineering,” and it’s why the child credit doesn’t fit the definition. The goal of the policy isn’t to get people to have kids they don’t want to have absent a credit; the policy is instead designed, as you note in your second paragraph, to offset certain distortionary features of the welfare state.
We’ve done a round on your but-kids-will-get-benefits-too argument before. Since you’re happy to repeat yourself, I’ll do the same. (The argument has also been answered by economist Hans-Werner Sinn.) One way of seeing the problem with your argument is to imagine a society with old-age entitlements much like our own but where the birth rate temporarily spikes and then reverts to its normal level. That spike permanently improves the programs’ finances: The ratio of contributors to beneficiaries temporarily rises and then falls, but not to below where it had been originally. Taxpayers with more kids contribute more to the entitlements than taxpayers with fewer kids.
A second way to see the flaw is to look at the entitlement programs as a way of socializing an age-old generational bargain. It is true that any kids you have will both contribute to your peers’ retirements and one day collect benefits of their own. But that’s not the full picture. You will get those future benefits only if your peers have kids, and your kids will get those benefits only if their peers have kids. Because raising kids is not costless, the bargain is substantially better for those with few or no children.
Third, we can look at the studies. If extra kids had the effect on old-age entitlements that you claim, we would expect them to show that expanding them raises birth rates. They show the reverse. That anti-child effect may not be an intentional program of social engineering by the federal government, but it is a large one–and one that ought to be shrunk or eliminated.
These arguments do not entail reducing parenthood to “a federal cash-flow question”: There’s a strawman if ever there was one. They just entail thinking about the impact of public policies on family structure, which it is folly not to do.