How does it look when we compare
money added to the economy by increases in credit market debt
to
money drained out of the economy as interest payments
?
Money added relative to money drained:

Graph #1: Annual Change in Total Debt, relative to Annual Interest Payments 
In the early years it's a wash, about oneforone. After the mid1970s it drops some: Less money is added than drained. Then there is a lastgasp peak between 2000 and the crisis, followed by sharp drop and the crisis. And now it is back in the normal range, as if nothing happened. Funny how that works.
A horizontal line at 1.0 would mean our new borrowing is equal to the amount we turn over to finance as interest payments every year. No effect on the money supply. Above 1.0 would mean we are borrowing more than we are paying as interest (increasing the money supply). And below 1.0 would mean we are borrowing less than we are paying as interest (decreasing the money supply).
What I want to see with this graph is how much of a boost borrowing gives the economy, and how much is only a boost to the financial sector. The graph is not a perfect indicator, because some portion of interest paid is withdrawn by the recipients and spent back into the economy. I don't know how much of it is withdrawn and spent, but it has to be somewhere between the zero line and the blue line.
Total debt is a "stock" but the change in total debt is a "flow". Interest paid is also a flow. The flowtoflow ratio is ... dunno, a ratio, I guess, as "billions per year" cancels "billions per year". Whatever. I want to look at the accumulation of differences over the years.
By "accumulation of differences" I mean the number shown on the graph, minus 1, and the sum of those values over time. Why "minus 1"? Because the graph shows the change in debt
per dollar of interest paid. I want to subtract the dollar of interest. If the blue line is at 1.1, it means we borrowed $1.10 for every dollar of interest paid. I subtract the dollar of interest paid to see how much economic boost we got from the borrowing: ten cents.
But if the blue line is at 0.8, it means that for every dollar of interest paid, we borrowed 80 cents. Subtract the dollar, and it turns out we're 20 cents short. The net effect of these financial changes was to create a drag on economic activity, rather than a boost.
(These examples assume that no money is ever spent out of savings, which is unrealistic. But the graph gives us a feel for what's happening, and a way to think about it.)
I determine the amount of boost or drag by subtracting the interest number from the changeindebt number. Then I add up the results to see the cumulative boost or drag. It's an interesting detail, an interesting indicator.
I took the FRED data from Graph #1 and did the subtraction and accumulation in Excel.

Graph #2: At 100% accumulation of Interest Received 
If no interest income is ever spent out of savings (100% accumulation), there is still some benefit to the economy from borrowing in the 1960s. But it goes negative in the 1970s, meaning there is more benefit to the financial sector than to the economy as a whole. And it goes negative rapidly, beginning in the late 1970s.
By the last year shown (2015), at 100% accumulation, the accumulated reduction of circulating money comes to ten dollars for every dollar of interest paid. I think that's an unrealistic number. I think it must be true that less than 100% of interest received is retained as savings, and that some portion of interest received is spent back into circulation.
Here is how the graph looks with a 50% accumulation rate:

Graph #3 
Now we see about $5 negative effect instead of $10. $5 still strikes me as unrealistic. But the more interesting thing about this graph is that the shape of the line is the same as before. The line still goes negative in 1970, and the downtrend still accelerates in the late 1970s.
If we cut the rate of accumulation in half again, the shape of the line remains unchanged:

Graph #4 
It's just more difficult to see. And if we reduce the rate to just 10%, meaning 90% if interest income is withdrawn from savings and spent back into the economy, the blue line is more compressed but again its essential shape remains unchanged:

Graph #5 
Can't see it now, but the benefit to the economy goes negative (becoming a drag on the economy) in 1970, and the downtrend accelerates in the late 1970s.
On this graph, the accumulated reduction of circulating money as of 2015 comes to about a dollar for every dollar of interest paid. Just a gut feel, that number is unrealistically low.
More on this
tomorrow eventually.