Why Capacity Utilization Matters

Had some discussion with JzB about Keynesian principles recently, what they are, whether they worked, whether they've been tried. The next day I saw An End To Keynesian Economics over at Maxed Out Mama, and it caught my eye.

The post opens with an analysis of the cause of depressions that I wish I could have written myself:

Depressions used to be called "panics", because depressions always incorporated a factor of drastic credit term changes. They would be preceded by sharp growths in lending, and then the end to a business cycle (aka recession) would produce a small change in loan quality. The small change in loan quality would act on the massive body of outstanding loans to change an ordinary correction into a panic, which in modern-day terms would be a credit contraction. So a depression usually was caused by a small real-world change which occurred at the end of a cycle of massive credit growth.

A small drop in loan quality acting on a massive body of outstanding loans. You know: excessive debt... excessive credit use.


The post provides this graph of wages and salaries...

...showing percent-change trends for 18-month and shorter periods. The uppermost line (the green line, showing the biggest changes) displays the 18-month changes. That is my focus here.

See the straight, downsloping trend line? Imagine sliding that line up til it touches the peaks of those three big green humps. What you see then is, at the rightmost end, the 18-month line does not reach up to touch the downsloping trend. Also, between the first hump and the last two, there are two smaller humps that also do not reach up to the downsloping trend through the peaks.

Does that remind you of anything? Reminds me of this:

In mine of 18 February I took the FRED "Total Capacity Utilization" graph and added red lines to show that "the high points have come at progressively lower levels." I also said

There are two exceptionally low peaks in the 1980s that I did not mark with red lines, as they do not fit the down-stepping trend. These two severe lows are associated with the Volcker squeeze.

When I saw the same pattern in Maxed Out Mama's graph I recognized it immediately.


The MaxedOutMama post includes a link to the numbers, even including which row in the file. Nice touch. So I thought I'd make my own graph. Even though I can't duplicate that nifty blue background.

Let me repeat the caveats that applied to the Maxed Out Mama graph:

This graph shows nominal seasonally adjusted payments to workers in the private sector. It does not include payments to government workers and it does not include benefits, such as medical insurance.

Mostly harmless, I guess. Next, I recreated the Capacity Utilization chart:

Each series graphed fine by itself, but there were a few problems putting the two data series on the same graph. For one thing, the Capacity Utilization numbers are monthly, and the Private Wage numbers are quarterly. Thought about it a while, then I came up with a formula and copied it down the column to "spread out" the quarterly data and make the dates match up with the monthly numbers.

Also, the CAPUT numbers pretty much range between 70 and 90 percent, but the Wage numbers range mostly between zero and 20 percent. So I added the constant 70 to each of the Wage numbers to push the wage trend-line up near the Capacity Utilization numbers to make visual comparison easier.

And for some reason, Google Docs kept wanting me to use a graph style that I never use. When I switched to the style I normally use, only one of the two trend-lines would show up. I think the two blank rows between quarter-numbers might be the problem.

I ended up using the graph style that Google Docs wanted me to use for the combined data. The two trend-lines match up extremely well:

Why does Capacity Utilization matter? Because wages and salaries rise and fall in sync with it. Need a raise? All you have to do is the impossible -- make Capacity Utilization go up.

And then wait 18 months.

//

The Google Docs spreadsheet for these graphs is accessible.