Taking a step back from the Excel FAIL heard ’round the world, it should be clear to most reasonable people that (Sum of Princpal of Debt Held By Public)/(Gross Economic Product) is not a very good way to understand the actual impact of government debt on socities. Even waving away the form of the debt (ie, whether it’s denominated in a currency the state has the authority to print), it’s very clear that various circumstances relating to monetary policy, trend growth, and structural factors are vital to understanding the impact of the debt-load on society. Karl Smith made this point excellently the other day, and springboarding off of that, I have a proposal. Someone – not me, because I’m busy, dammit – should create an index of the present value of all future interest payments owed by the government divided by the present value of NGDP into perpetuity. This is easy if you consider government debt a kind of perpetuity in which the principal is continually rolled over until it shrinks to a vanishingly small amount of NGDP, which it is. Therefore, the true burden of debt is all the interest owed on it, forever. That number should be easy to project as long as you know the current coupon amount and rollover date for all government debt, as well as have a rough estimate for what the prevailing interest rate will be whenever any given bond is rolled over. This is probably pretty easy to guesstimate without actually knowing the data to that fine a degree. The best way to do this would be to index future expected interest rates to future expected NGDP (since they are closely linked), and you could then see the actual share of future economic output that will have to be paid out to bondholders.

Or you could have a Manhattan. I like Manhattans.

Inline image 1