Friday, July 8, 2011

Relative to NGDP


This morning I showed this graph:


Clonal said:
Redo the same graph as a percentage of nominal GDP. It is much more interesting. The deficit goes up in every recession. The only difference is that this last recession was a real doozy.

This is because, without additional govt spending, the tax receipts go down in a recession. If govt. spending increases, the tax receipts may not go down. However austerity, and cutting govt. spending will not (with 100% certainty) make the deficit come down as a percentage of GDP.

First of all,


There ya go. More interesting? Dunno. But I'll poke it with a stick.

General downtrend, until 1982-83. Then up, up, a weak up, and oops! Down she goes.

Second, "The deficit goes up in every recession."

For those who need to be coddled: When you're talking about deficits, down is up and up is down. If you take a magic marker and use it to highlight downtrends near recession bars, you find a lot of 'em.

You're right about that, Clonal. And I do agree with your explanation of it. And I hope some day people will expect to find the correlation you point out.

Here's mine, with highlighting.

Third, regarding your version of the graph:
Already by the 1954 recession, the area above the curve (up to the zero-line) is bigger than the previous area below the curve (down to the zero-line). So already the deficits were bigger than the surpluses. And from there it was all downhill, until 1983.

Why?

Then it went up.

Why?

And then it went down again.

Why?

So much to talk about.

5 comments:

Clonal said...

If you do the same , but substitute real GDP, the "relationships" that you noted, disappear. Why?

Think back to debt. Debt and interest are repayable in nominal $ not in real $. Looking at the monetary flows may be useful.

Clonal said...

And of course, taxes are paid in nominal $, and government and private spending is all done in nominal $. Wages are paid in nominal $. So as far as the economy is concerned, "real" or inflation adjusted statistics should only be used when specifically needed to answer a specific question that is related to inflation and buying power. Otherwise they will mask or hide the impact of the monetary flows.

The Arthurian said...

Clonal: "government and private spending is all done in nominal $."

This is absolutely true, and it is neglected by real economists in ways that we -- I want to say nominal economists, but I won't -- do not dismiss so lightly. (I'm no economist. I don't know about you.)

In Money Mischief, Milton Friedman's famous "money relative to output" graphs use as a numerator M2 -- which includes money in the spending stream AND money NOT in the spending stream. Unjustifiable.

As if that isn't bad enough, his graphs use as a denominator "real" GDP, which is nominal or actual GDP with inflation factored out of the numbers.

The way Friedman uses it, as a denominator, real GDP factors inflation IN to his results.

In other words, his graphs show that 'money relative to output' (MRTO) looks like the price level, because he factors the price level into his numbers. This is akin to fraud.

I might say, economists should ALWAYS work in nominals. And then to check their results, they can compare real GDP to real GDP over time. And then go back to the nominals, the actual numbers.

Come to think of it, the real economists are the ones who are economists in name only.

Clonal said...

Rodger Mitchell had an interesting post today - – Why bank lending leads to recessions. A counter-intuitive finding.

He comes to a conclusion that Recessions follow increases in state, local and private debt

Clonal said...

Also, the three asset bubbles are clearly marked in the graph - S&L, dot com and the housing/RE bubbles. That is all in the post 1983 period.