Being the most charitable I can be

In this chart, I once again have total debt costs computed as debt payments divided by public debt, per period.  The read line is real growth smoothed with annual compounding.  I think exponential assumption causes phase shift. I use this method, it gives actions done by government in nominal money, and measures real growth with some sense of forward planning. So Blue above red is a money loser, blue below red is a money winner.  When blue is near the volatility of read we are undefined.

Of course, Reagan was the big loser, and George I had to deal with aftermath. Clinton did the best, George II got a bit of bubble. Obama has been in the mud. And the government effective interest is a good smoothing filter for growth, going down.

Effective guv interest rate has inertia.  It takes four years to roll the debt over, so lower or higher interest take time to reach the expense account.  So a growth surge generates  private sector losses.

Is growth getting lower, government more efficient, inflation more managed? Why have effective rates dropped? Not seen on this chart.  The red is exponentially smoothed, the blue not; so this is DSGE to current sample; apples to oranges.  Compare them as current samples, current nominal values, period by period to see it as they saw it. I think you will find the ten year yield is the low pass filter for the current growth. See if it leads of lags and you have your culprit, say Detective Holmes.