Saturday, September 15, 2012

Two ways to look at debt


My "sales pitch" for my "Measuring the Erosion of Debt" pdf included this line:

If there is such a thing as the erosion of debt by inflation, then inflation-adjusted debt on a graph must be higher than nominal debt.

There was not a doubt in my mind. (That was 17 August)

A few days later I read Thayer Watkins' page on The Deficit of the Government. Watkins wrote:

This inflation adjustment is especially important during times of high inflation that push up interest rates and require large interest payments on the debt. The real value of the deficit would be far smaller than the nominal value.

//

Wow.

Here's part of what I have for tomorrow's post:

When you show real and nominal output on the same graph, the inflation-adjusted line always crosses the nominal line at the base year. And if, as a rule, there is always inflation, then before the base year, the inflation-adjusted line is always higher. And after the base year, it is always lower. The year we pick for the base determines where the lines cross, where the adjusted numbers are higher, and where the adjusted numbers are lower.

So then Watkins is right (the real value is smaller than the nominal) if we are thinking of a year that is *after* the base year. For example, the base year is any year in the past, and we are thinking about what we owe now, this year.

"What we owe now" is a useful way to look at debt, for sure. However, if there is such a thing as the erosion of debt by inflation, then inflation-adjusted debt on a graph must be higher than nominal debt.

If there is confusion here, it arises from the confusion of stocks and flows. Inflation affects stocks and flows differently.

That's all for now.

10 comments:

Anonymous said...

I do not understand. Can you reword?

The Arthurian said...

No, I don't understand it either. Thought I did, but at the last minute ended up erasing most of what I had written. (You caught me!)

I bought a house in the 1970s. And with the inflation of that time my paycheck kept going up. It got easier and easier to pay the mortgage. So I know from personal experience that inflation erodes debt.

Inflation -- assuming it works its way into wages and not just prices -- reduces the burden of existing debt.

If we rejigger the numbers to imagine that there was no inflation, my income would not have gone up so much. My mortgage payment would have continued to take more of that income. Adjusted for inflation, my debt is higher. Therefore I say: inflation-adjusted debt on a graph must be higher than nominal debt.

The rule for GDP or output is different, because GDP starts from scratch every year. Lots of graphs tomorrow.

Thanks.

Clonal said...

Art,

You might also be interested in reading David Graeber's Can Debt Spark a Revolution?

Quote:
When we were organizing the Wall Street occupation in August of 2011, we really didn’t have any clear idea who, if anyone, would actually show up. But almost immediately we noticed a pattern. The overwhelming majority of Occupiers were, in one way or another, refugees of the American debt system. At first, that meant student debt: the typical complaint was “I worked hard and played by the rules, and now I can’t find a job to pay my student loans—while the financial criminals who trashed the economy got themselves bailed out.”

What was remarkable wasn’t so much the fact that the camp began to fill with so many debt refugees, but how much their plea resonated across the political spectrum. In the 1960s or early ’80s, the plight of a college graduate juggling loans wasn’t the sort of thing most likely to wring the hearts of transit or sanitation workers. But Occupy received warmth and solidarity from organized labor. Something clearly had changed. We had come to see ourselves as members of the same indebted class.

Jazzbumpa said...

I won't be around tomorrow, and will try to catch up later in the week.

Inflation is always relative to a base year that is arbitrarily chosen in the extant moment or the past. Constant 2000, 2005, 2009, etc $$$.

I can almost see real debt being higher than nominal debt in the past on a graph, since inflation reduces the value of the debt.

But something is bothering me that I can't quite pin down.

JzB

Jazzbumpa said...

It seems to me that the right way to adjust debt for inflation ought to go something like this.

Since debt is a stock and deficit is a flow, each year the nominal debt increases by the amount of the current deficit.

To start, take debt in some past year long enough ago that the number is small relative to the current number, say 1950.

For 1951, adjust the existing stock of 1950 debt by the 1951 inflation number (CPI or GDPDEF?)

Add to that the deficit flow for 1951, adjusted by 1/2 of the 1951 inflation number. This assumes equal inflation along the course of the year. More exact would be to add each month's (or quarter's) deficit, adjusted by the fraction of the year it was in the stock, 11/12 for Jan, 3/4 for Q1, etc.

But the mo or Q protocol is probably not worth all the extra work; and I don't know if data that fine-grained is available.

Either way, this gives you the 1951 stock.

Adjust that by the 1952 inflation no. and then add the fractionally adjusted 1952 flows.

This gives the 1953 stock.

Etc.

Does that make sense?

JzB

The Arthurian said...

YES!! Well said.

1. I have never talked about the granularity of the numbers. I thought it would be too confusing to add that complicating factor. But yes, if you are breaking up total debt into yearly sums, why stop there? Debt and the deflator are reported quarterly; the CPI, monthly.

Not only the explanation, but also the arithmetic is more complicated by using quarterly data. But I'm sure it could be done that way.

If you only adjust for half the 1951 inflation in 1951, what happens in 1952? Do you adjust for the other half of 1951, plus the first half of 1952? I can see the sense of this, but it seems like an unnecessary complication. (However, as I have not done graphs that way, I do not know if it is unnecessary or not.)

CPI is for consumers, and GDPDEF is for GDP. I think GDPDEF is more appropriate for general use.

2. Your 1951 result has 1951 for a base year. Your 1952 result has 1952 for a base year. What you have here is a floating base year. This is interesting and possibly very useful. I have not done it that way.

3. As you say, one starts with a lump of debt that cannot be broken into prior-year subtotals (perhaps because there are no prior-year numbers to look at) but this amounts to a small error, I agree. Because that number is small and distant. Anyway, the trend SINCE that first year provides most of the information we need to see.

4. When adjusted as you and I are thinking, the "real" debt numbers differ significantly from "real" debt numbers calculated from the running total. Therefore the ratio of debt-to-gdp, stock-to-flow, is different also.

In other words, the picture of debt is different because of the way we are calculating it. And the picture of debt must affect the analysis of the economic problem. So this new picture of debt is significant.

Thanks Jazz. See you in a few.

Jazzbumpa said...

If you only adjust for half the 1951 inflation in 1951, what happens in 1952? Do you adjust for the other half of 1951, plus the first half of 1952? I can see the sense of this, but it seems like an unnecessary complication. (However, as I have not done graphs that way, I do not know if it is unnecessary or not.)

This is simply a way of approximating the finer granularity with annual data.

Debt from early in the year gets the full year's inflation, from later gets little to none, half is the average.

Off to T-town. Gotta run.

Cheers!
JzB

Greg said...

One comment;

You were looking at your private debt situation in your opening comment Art and you made the valid point of higher income allowing easier pay down of old private debt that stays nominally the same. So as long as inflation translates to more income then its actually a benefit for some.

Now youve gone and started talking about govt debt and how inflation affects it. Im not sure one can use the same analysis since they are of opposite types. Relative to the citizen, govt debt is an asset while private debt is a liability. I can actually use my 100,00$ govt bond to pay off my 100,000$ mortgage. And if Im getting 5% on that bond I can pay it down quicker.

Another thing very different about my private debt is that I can always just go and pay down the principle and avoid interest costs. There is no such equivalent scenario with govt debt. Once a bond is issued at a certain term at a certain interest rate the govt will and must pay that interest til maturity. It can try to "buy" the bond back (QE 1,2,3) but it can never just go and say "Here's your 100,000$ back we are paid off".

So the costs of interest are perpetual with the govt bonds, but they are usually miniscule in regards to the principle which is generating them. (When the govt issues me a 100,000$ bond I take 100,000$ of buying power today and exchange it for 400$/mo of buying power for the term of the bond, assuming 5% interest).

In terms of its affect on inflation, as most people talk about inflation, not having 100,000$ to spend today but instead getting 400$/mo should be less inflationary.

I dont think one should try and think about govt and private debts using the same model. It takes you places that are contradictory, since they are on opposite ends of the spectrum in accounting. How inflation affects assets vs liabilities is different. I want inflated assets and deflated liabilties.... ALWAYS!

The Arthurian said...

But apparently, you can't use your government bond to pay off your mortgage until the bond comes due... :)

Debt is an asset to the creditor and a liability to the debtor, citizenship no matter, public or private no matter. Government debt is an asset to the bond-holder, but not to his footservant.

The guy making 400$/mo on his 100,000$ has no plans to spend the 100,000$. That's the thing Bernanke is trying to change.

There's too much money already in the hands of people with no plans to spend it. And Bernanke is giving those people more money.

Greg said...

"But apparently, you can't use your government bond to pay off your mortgage until the bond comes due... :)"

Well the interest payment can always be used to pay it down and I believe you can always sell your bond back quite easily if all you want is your principle.



"Debt is an asset to the creditor and a liability to the debtor, citizenship no matter, public or private no matter. Government debt is an asset to the bond-holder, but not to his footservant."

Not sure what your saying with your last sentence. Im guessing you are talking about the distribution of govt debt, ie that while its true that govt debt is a private sector asset its not an asset for everyone in the private sector just the few bondhlders. And this is certainly true but most anyone with an insurance policy or any kind of pension/retirement plan is also a partial owner of govt debt as well since safe bonds form an important income base for portions of almost all retirement plans and insurance policies.

I however would prefer to see our counties national debt payments go out to all citizens in the form of a citizens income


"The guy making 400$/mo on his 100,000$ has no plans to spend the 100,000$. That's the thing Bernanke is trying to change."

Agreed, but he is no more likely to put it to productive use just because you take away his small but safe interest. He's likely to find the next safest investment he can find and not necessarily put it to new ventures and job creation.


"There's too much money already in the hands of people with no plans to spend it. And Bernanke is giving those people more money."


Not sure what you mean by giving them more money. Making their savings acct (govt bond) into a checking acct (cash) only changes their risk profile, its not giving them any more money, but as you said above, he hopes/expects/intends for them to take the cash and buy something that will lead to new ventures and job creation. Its possibly worse than that though because I believe he (and the other monetarists) actually think that by buying even riskier assets and driving up those prices of stocks, commodities, houses they will feel richer and this "sense of wealth" will then remove anxiety and allow them to behave like a normal entrepreneur. Its all about restoring a feeling of security to the 1%ers who are responsible for the worlds well being and economic output.

Sad state of affairs for our best and brightest if you ask me