Continuing my meditation on Brad DeLong’s meditation on the fiscal future. Brad doesn’t just argue that governments should be bigger in the future; he also argues that governments have historically not had enough debt, and should have more.
Why? Because, he says, r-g — the difference between the real rate of interest on government debt and the rate of economic growth — has been consistently negative. Why is this significant?
Well, we normally imagine that if a government engages in deficit spending now, it will have to engage in compensating austerity of some form later — even if it doesn’t plan to pay of the debt, it will still have to cut spending or raise taxes so as to run a primary, non-interest surplus if it wants to stabilize the ratio of debt to GDP.
But when r is less than g, a higher debt stabilizes itself: erosion of the debt ratio by growth means that no primary surplus is needed. So you can eat your cake and have it too. A bigger debt lets the government do useful things, like invest in infrastructure; it gives investors the safe assets they want; and it need not lead to any future pain as long as you don’t do foolish things like join a currency union with no well-defined lender of last resort.
But is r really less than g for all major players? Brad uses the average interest rate on debt, which I haven’t had time to compute. What I’ve done is use the 10-year bond rate — which is somewhat higher than the rate Brad uses, I believe — and examine the G7 over the period 1993-2007. And I think we get some interesting insights.
First, all of the G7 paid roughly the same real interest rate, using GDP deflators to measure inflation:
As I’ve noted before, this doesn’t mean that the Wicksellian natural rate was the same everywhere; in the case of Japan, at least, the actual rate was well above the rate consistent with full employment. In any case, however, arbitrage looks quite strong.
However, countries differed a lot in their growth rates, so that r-g varies considerably. And this raises the question, did the “right” countries have a lot of debt?
Compare debt ratios in 2007 with r-g estimated over the 1993-2007 period:
For English-speaking members of the G7, r-g is slightly positive, but would be negative if I used a broader interest definition. But it was much higher in Japan, Italy, and Germany, which all had slow growth over this period — and Japan and Italy also had high debt. (The causation almost surely ran from slow growth to high debt, not the other way around.)
This suggests, I think, that Brad’s case for higher debt, while powerful, doesn’t apply to everyone. It’s a good case for English speaking members of the G7 and also for Germany looking forward (the 10-year index yield is -1 percent). Unfortunately, the biggest debt accumulations have come in economies that have much lower growth — mainly demography in Japan, productivity collapse in Italy.
No strong moral here, but I do think we need to be careful not to assume that the US case generalizes to everyone. Hellenization — assuming that we’re all Greece — has been a big problem in recent years; but Americanization — assuming that the US is representative — could be a problem too.