Friday, June 24, 2016

In a credit-addicted society, you need a hell of a lot of inflation to get debt going down


Came across an old (2011) Krugman post the other day. Here's the opening:
Richard Koo has another paper on balance sheet recession out (pdf), with good charts for a number of countries. I still have some differences with him over monetary policy — I still don’t understand why he doesn’t see debt-eroding inflation as something helpful in dealing with debt overhang — but his view of the sources of our Lesser Depression is completely right.

I heard of Richard Koo and the balance sheet recession. But I never read Koo and I know nothing about the B.S. recession. That's okay, maybe I'll read the paper Krugman linked if I'm still interested later.

Anyway, Krugman has "differences" (plural) with Koo, but identifies only one:

I still don’t understand why he doesn’t see debt-eroding inflation as something helpful in dealing with debt overhang

It's been five years. Hopefully there has been an answer by now. I'll tell you my answer first, then maybe go look for other answers.

//

Q: Why doesn't Richard Koo see erosion of debt by inflation as a way to reduce debt?

A: Oh, that's easy! In a credit-addicted society, you need a hell of a lot of inflation to get debt going down. And actually, inflation doesn't make debt go down. It only makes NGDP go up faster than debt ... unless the rate of debt growth goes up.

Here's the thing. Inflation -- as long as incomes keep up with it -- makes existing debt easier to bear. But inflation makes new additions to debt bigger. So it's not all neat and tidy.

//

I threw together a spreadsheet so you can experiment with debt growth and inflation rates. Here's a screen cap:


The red line is debt. The blue line is GDP, actual ("nominal") GDP. The yellow cells contain numbers you can change to change the graph. The NGDP level and the Debt level both start at 100, for convenience and to make comparison of changes easier. The rates of growth of debt and inflation-adjusted ("real") GDP and the rate in inflation, the rates shown are averages for the 1950-2015 period. (The FRED data is included in the file, and period averages are shown to the right of the graph.)

If I leave the starting levels unchanged and leave the growth rate of RGDP at 3.3% and Debt at 8.0%, then in order to get the blue line a bit above the red line -- in order to make the Debt-to-GDP ratio fall -- I have to increase the rate of inflation from 3.2% (shown) to 5%, as the following graph shows:

Graph #2: The Effect of 5% Inflation
Huh, that's less inflation than I thought. Still, at 5% inflation, prices double in less than 15 years. Quadruple in 30 years. If you ain't big on globalism, if you want a strong nation, you need the nation's money to hold its value. It's very simple really.

//

For the record, the patterns shown on Graph #1 are a pretty good match to the patterns of actual data shown of this FRED graph:

Graph #3: Actual (not calculated by me) Data for GDP (blue) and TCMDO Debt (red)
A pretty good match. The blue line on Graph #3 lags behind a little. That's because debt hinders growth, but that's off-topic today.

And for the record, the whole "inflation erodes debt" argument depends on the assumption that disposable incomes keep up with prices, an unreliable assumption.

And by the way, it isn't that inflation erodes debt. It is that inflation erodes the value of the dollar, which pushes up new spending but not old debt. I used to like Krugman's term "erosion". I'm starting to think it is as deceptive a term as "real" (as in "real GDP").

And anyway, inflation is not a solution. Inflation is a problem. The fact that inflation can sometimes help to keep debt down (relative to GDP) does not mean inflation isn't a problem.

No comments: