When they remove the inflation from (or "deflate") GDP, they get what they call "Real GDP". (I don't like to see the word "real" used that way, but what can you do? I'm just trying to go with the flow and get to the point!)
You have to be careful when you use Real GDP in a calculation. For example, it is not good arithmetic to add normal (inflating) numbers to deflated numbers, because that's apples and oranges.
It is not good arithmetic, either, to multiply
or divide an inflating number by a deflated one, because combining the
two types of numbers this way skews the result. For example, if we
divide the quantity of money by Real GDP, the result is skewed upward by
the inflation we leave in the money number. Shown on a graph, the apples-and-oranges
result will appear strikingly similar to the path of inflation. That was
the bad arithmetic Milton Friedman used for the "money relative to
output" graphs in his books Money Mischief and Free to Choose. The artificial similarity shown in his graphs made them most convincing. But it was not good arithmetic.
Setting such concerns aside, today I am going to combine inflating and deflated numbers. I want to divide Real GDP by the Federal Debt. Real GDP is deflated; the Federal Debt is not. Dividing by an inflating number will skew the result downward. But before we get to that...
Everybody looks at Debt-to-GDP using numbers that are not adjusted to remove the inflation. That's fine, I have no problem with it. I am only pointing out that the ratio is often used and often discussed.
Some people look at federal debt relative to GDP. Others use nonfinancial debt (which includes business and consumer debt along with federal, state and local government debt) relative to GDP. They differ, but both have their uses.
Today I will use just the federal debt relative to GDP -- but I will turn the ratio upside down and look at GDP relative to the federal debt. And I will use the deflated version of GDP, Real GDP, as noted above. And thus we come to the part where I am doing bad arithmetic.
But is it bad arithmetic? I'm not sure. And even if it is bad, sometimes you can get away with it when it is for a good reason, as with Friedman. Let me tell you what I have in mind. I'm thinking of Real GDP as a measure of the economy's performance. That is what it's used for. That's what people object to when they object to GDP, saying It doesn't measure happiness.
But anyway, just for today, let's not call it "Real GDP". For today, let's call it "Economic Performance".
This is the logic that tells me the arithmetic may be acceptable: I want to look at economic performance relative to the federal debt: How well does the economy do, as the federal debt grows in size?
The answer: Not well at all:
 |
| Graph #1: Economic Performance relative to the Federal Debt |
The
ratio peaks in 1941 and falls to 1946. After 1946, the ratio rises to a
peak in 1966 (shown by the first blue dot). But that was soon
after WWII, when "federal debt relative to GDP" was falling.
Graph #1 shows a gradual decline from 1966 to 1973 (the second blue dot), and then rapid decline to the early 1990s. There is slight increase in the latter 1990s, and then the decline resumes.
I never did find a statement from Minsky where he says the "golden age" ended in 1966 -- Steve Keen
mentions "what Minsky described as the 'robust financial society' that
underpinned the Golden Age that ended in 1966" -- but you can see it on
the graph: the rising economic performance changes to falling economic
performance at the 1966 peak.
The end of the golden age is more commonly dated at 1973. You can see this on the graph, too, as economic performance falls rapidly after 1973. In the immortal words of Scott Sumner:
I am not denying that growth in US living standards slowed after 1973, rather I am arguing that it would have slowed more had we not reformed our economy.
Sumner confirms the 1973 date. He also sees "America's amazing success since 1980". But then, he wasn't looking at Graph #1.
I'm tempted to say the ratio is unfair, not because Graph #1 combines deflated with inflating numbers, but because GDP starts off every year at zero, while debt starts off every year where it left off at the end of the year before. That sort of accounting would skew any graph showing a debt-to-GDP (or GDP-to-debt) ratio. The only way to avoid that skewing would be to have debt completely paid off, every year, so that debt also starts off every year at zero. But that would be a terribly unworkable constraint to impose on the economy.
We
need better economic performance, but rapid increase in the federal debt
has clearly not delivered. With the ratio rising in the 1950s and early
'60s, and running then about 10 times as high as it has been in recent
years, it is obvious that the increase of federal debt since 1974
has been extremely ineffective in boosting economic performance. And
increasing the federal debt even faster would drive the Economic
Performance ratio down even more!
Thus I say it cannot be that the solution to our economic troubles is to increase the federal debt even more rapidly than we have done and "accept whatever deficit may result." That cannot be the solution. It will not work. It has not worked. It can not work.
Consider the trend of federal debt growth during the years 1946 to 1974, the red line on the graph below:
 |
| Graph #2: Gross Federal Debt in Billions, and the 1946-1974 Trend |
By
the late '70s the federal debt was growing faster than the post-war trend. It
has since raced ahead. Had it stayed on trend, by the year 2019 the
federal debt would have been less than $2000 billion. But it didn't stay
on trend, as the blue line shows, and by 2019 the federal debt was more
than $22000 billion.
In 45 years, the federal debt increased more than tenfold above what it would have been had it stayed on trend.
That
increase pushed the ratio down, as shown on Graph #1. NOTE CAREFULLY: I
am not saying that the growth of federal debt "caused" a decline in
economic performance and Real GDP growth. By the arithmetic, however, the rapid increase in federal debt
forces the Economic Performance ratio down more and more. Call
it bad arithmetic, if you like. But it is no
coincidence that the rapid decline on Graph #1 begins in the mid-1970s. (H/T Milton Friedman)
Accelerating the growth of federal debt has not boosted Real GDP enough that the Economic Performance ratio suggests healthy or even normal performance. It would be fair to conclude that the under-performance of Real GDP is not due to insufficient increase in the federal debt.
Some people attribute that under-performance to excessive increase in the federal debt, but my graph shows no evidence of it. My graph does not show that the rising federal debt caused Real GDP to under-perform. It only shows that the rising federal debt failed to prevent that under-performance.
Note,
however, that we have been looking at Real GDP in relation to federal
debt alone -- and that the federal debt has been the smaller half of
debt since 1952. The bigger half is debt other than federal,
and most of that is private-sector debt. Our debt. The graph and the
ratio and the paragraphs above do not consider private debt. But we have
to consider it.
Steve Keen says we need some kind of private-sector-debt forgiveness. That would work like magic, I think, to improve economic performance. But I also think the Federal Reserve has to stop allowing debt to grow more rapidly than the quantity of non-interest-bearing money. It has been Fed policy to allow (and to encourage) the growth of debt at a more rapid rate than the quantity of money, perhaps because they think increasing the quantity of money causes inflation but increasing the quantity of debt does not.
That, however, would be flawed logic on their part.
No comments:
Post a Comment