More On My "Deflationary Collapse" Ticker
Courtesy of Karl Denninger at The Market Ticker
Boy, that got people’s attention.
We must change our economic course now and accept the contraction that MUST COME in order to save our economic and monetary system.
Yep.
It also drew critics, chief among them being people who claimed that "debt isn’t really growing at that percentage beyond GDP."
Oh yes it is.
In fact, I was being rather reasonable, in that I was using time periods going all the way back to the 1950s.
But – let’s focus in on "closer to today", specifically, 1990 forward.
I think we can all agree that the 1990s were tremendously good times for GDP, and were not nearly so destructive for debt, right?
Ok.
Well, I hate people telling me they think I’m being unrealistically "doom and gloomish", so I "de-theoreticalized" that chart a bit, and picked on more-recent times where the growth rates were pretty darn good and some of the really-expansive debt periods, particularly from the 70s, were not included.
Here are the ugly facts – not conjecture, not assumptions, not going back 50 years data, current and unassailable facts.
During the period from 1990 onward, GDP grew at a compound annual rate of 5.361%1. Debt during the same period, nearly 20 years, grew at a compound annual rate of 7.9401%.
It is worse if you look at 2000 onward – there GDP growth was 5.25% on an average annual basis, while debt growth was 8.6279%.
I thus have "recast" the 20-year forward graph for you in two forms – one assuming we can maintain the 1990 onward rates (which are more favorable) and the second assuming that the 2000 onward rates are what we face going forward.
Note that the GDP growth rate exceeds (by at least one percent and in most cases two percent!) the expected actual economic growth rate for at least the next five years from nearly all mainstream economists. I am also not building in any extraordinary increase in debt for the projected (by both the White House and CBO) increase in Federal Debt by $9 trillion over 10 years – a 100% increase – as it roughly corresponds with the expected debt increase in the economy as measured by previous trends. Nor am I including any of the unfunded liabilities of Social Security and Medicare, which total some $70+ trillion dollars all on their own. Since my intent is to discover whether there is any rational basis for a belief that we can manage to be ok given the growth in debt and GDP projected, I am willing to use wildly optimistic GDP numbers (compared to those of all mainstream economists) in these projections forward.
Both the 90 onward and 2000 onward time frames included really big "booms", Internet and Housing, respectively. This in no small part is why there’s a problem with growth forward and I simply don’t believe we will be able to maintain even a 3% compound growth rate over the next 20 years, but I’ll go ahead and use the historical averages from 1990 onward and 2000 onward and grant the naysayers that we can conjure up another big speculative boom.
Again, this assumes no further recessions for the next 20 years (hah!) and thus no interruption in growth.
Here are the graphs:
Even with these outrageously optimistic assumptions we still fail, as debt rises to anywhere from 237% to 275% beyond today’s levels while GDP only increases by 130% in the best case, and 100% in the somewhat-more-realistic case.
That is, debt service costs are going to double or even triple compared to GDP.
To be blunt: No they won’t.
They can’t.
And remember folks, I’m being terribly optimistic here.
Why do I have zero faith in the possibility of getting this under control unless radical policy changes are made? This graph right here:
I call this my "Ponzi Finance Indicator"; it is quite simply the percentage difference between the annualized rate of change in GDP and debt.
Negative values indicate that debt is growing faster than GDP; positive values indicate GDP is growing faster than debt. The former is negative for the economic outlook, the latter positive.
When debt is used as self-liquidating trade credit or for other productive purposes then this indicator will trend above zero, as GDP growth is driven to a greater degree as a percentage change than is debt.
When debt is used for purely speculative or consumptive purposes this indicator will trend below zero, indicating the non-productive (and ultimately destructive) use of debt.
When you get higher lows as we had from the 1950s to roughly 1971 the economy is growing solidly and debt is not a problem – it is being used properly to advance GDP through self-liquidating trade credit, the purchase of productive machinery and the like.
But when you see a general deteriorating trend as we have had since the mid 1970s you’re in trouble. Since 1980 we have had only one short period where GDP growth outpaced debt growth on a percentage basis – that is, there was only one year out of the last twenty-nine where this was the case.
This graph, by the way, goes a long way toward destroying the "Laffer Curve" argument. Yes, tax revenues rise when tax rates fall (to a point; obviously a zero tax rate generates no revenue) but that is more than offset by behavior in the general economy where decreases in tax rates spur more and more borrowing. This in turn depresses the "Ponzi Indicator" and taken as a whole the "Ponzi Indicator" determines whether the economy is stable – that is, whether GDP is growing faster than debt in the economy as a whole on a relative basis.
A deficit in the Ponzi Indicator cannot continue indefinitely, yet we seem to be oblivious to the reality of what is coming if we keep attempting to steadily march on this path going forward.
The charts above show you what’s in store for us if we do not change course, and given the continual "lower highs", now not breaking the zero line since 1993, we’re in extremely serious trouble.
1 Methodology note: To compute annualized growth rates the GDP and Debt sampling numbers were taken from the BEA and Federal Reserve Z1 numbers, and each quarter’s annualized growth rate was computed as the change from the identical quarter one year prior, expressed as a percentage. These samples were then summed and divided by the number of samples to arrive at the average compound growth rate over the period. This "de-noises" the numbers compared with annualizing quarterly changes by multiplying the quarter-over-quarter change by four.