Which one is not like the others?

Earlier this week I showed the shocking decline of Fixed Private Investment relative to total accumulated debt:

Graph #1: Fixed Private Investment relative to Total Debt
It shows significant changes before the 1980 recession, after the 1990 recession,
before the 2001 recession, and toward the end of the 2007-2009 recession.

Today, we look at Fixed Private Investment relative to components of the total debt:

Graph #2: Fixed Private Investment relative to Nonfinancial Debt
It shows significant changes before the 1980 recession, after the 1990 recession,
before the 2001 recession, and toward the end of the 2007-2009 recession.

Graph #3: Fixed Private Investment relative to the Publicly Held part of Federal Debt
It shows significant changes before the 1980 recession, after the 1990 recession,
before the 2001 recession, and toward the end of the 2007-2009 recession.

Graph #4: Fixed Private Investment relative to Non-Federal Non-Financial Debt
It shows significant changes before the 1980 recession, after the 1990 recession,
before the 2001 recession, and toward the end of the 2007-2009 recession.

Graph #5: Fixed Private Investment relative to Household Debt
It shows significant changes before the 1980 recession, after the 1990 recession,
before the 2001 recession, and toward the end of the 2007-2009 recession.

Graph #6: Fixed Private Investment relative to Financial Debt
It seems to show a trend change toward the end of the 2007-2009 recession.

Which one is not like the others?

Log of Real Fixed Private Investment (and Trends)

Hover the mouse over the graph to see the trend lines I sketched in...

Graph #1: Fixed Private Investment, Deflated, Log of it.

Looks like investment growth trended at one rate right up to 1980, and at a slower rate thereafter. I don't imagine I'm the first guy to say that. But I do want to point out that the growth of Real GDP shows a similar slowdown right around 1980.

Of course. It makes sense that if economic growth slowed in 1980, the driver of economic growth also slowed around that time. It makes sense. The thing is, nobody says it. No. For example, Scott Sumner says this:

growth was slowing almost everywhere in the 1970s and 1980s

and this:

growth in US living standards slowed after 1973

After 1973? Yeah, about seven years after 1973.

Between the 1974 recession and the 1980 recession, Graph #1 shows FPI increasing faster than at any other time since 1960. Growth was strong after the 1974 recession. Growth didn't slow, then. Growth didn't slow until Paul Volcker suppressed inflation and disposed of the corpse of vigor.

Just by eye

Continuing our look at Fixed Private Investment and Total Debt. The growth of fixed private investment peaks before the growth of total debt:

Graph #1: Fixed Private Investment (blue) and Total Credit Market Debt (red) on Log Scales

The values are shown on a log scale to make growth rates visually evident.

Investment (the blue line) appears to curve gradually upward until 1980, then trends upward at a slower pace, on a straight rather than upcurving path.

Debt (the red line) appears to curve gradually upward until around 1986, then trends upward at a slower pace on a path that displays little if any upward curve.

The upward curve of the blue line appears to run from 1960 to 1980; that of the red line from 1970 to near 1990. Either (or both) of these upcurves might be related to "the Great Inflation" -- and they probably are. But it is difficult "by eye" to imagine why inflation would appear first in the investment numbers, and later in the debt numbers.

A Barometer of Growth

A search for PFI at FRED turned up plenty of results but only one containing the word "investment":

Purchasing Power Parity Converted GDP Per Capita (Laspeyres), derived from growth rates of Consumption, Government Consumption, Investment for Finland

I didn't go there.

That search also turned up one result containing the word "private":

Nonfarm Private Financial Activities Payroll Employment

I didn't go there.

It turned up no result containing the word "fixed".

A search for FPI at FRED brings you immediately to the Fixed Private Investment graph. There, the notes tell us

Notes:

BEA Account Code: A007RC1

A Guide to the National Income and Product Accounts of the United States (NIPA) -
(http://www.bea.gov/national/pdf/nipaguid.pdf)

At the nipaguid PDF -- a name that satisfies the 8-character limit of MS-DOS file names, in case you haven't yet upgraded -- we find the phrase private fixed investment on page 5 and in a table on page 7, and on page 14:

Gross domestic investment (6-1) measures the total investment in the United States in fixed assets (that is, the structures, equipment, and software that are used in production) and in inventories (change in private inventories). It is the sum of private fixed investment (see 1-20), government fixed investment (see 1-29), and change in private inventories (1-25).

I quit looking for PFI at that point. I like that definition. It reminds me of something I once knew: that change-in-inventory counts as part of investment. That's an important bit of trivia which explains the "S=I" thing people are always talking about. Not relevant to this post, but worth remembering, certainly.

There were no occurrences in the nipaguid of the phrase "fixed private investment".

So what FRED calls FPI and what BEA calls PFI are the same thing. I can talk of Fixed Private Investment and Private Fixed Investment interchangeably, with impunity.

Good. Because in a different PDF from BEA -- or actually in Chapter 6 of what looks like the same nipaguid, but in a different file -- we find this:

Private fixed investment (PFI) measures spending by private businesses, nonprofit institutions, and households on fixed assets in the U.S. economy. Fixed assets consist of structures, equipment, and software that are used in the production of goods and services. PFI encompasses the creation of new productive assets, the improvement of existing assets, and the replacement of worn out or obsolete assets.

The PFI estimates serve as an indicator of the willingness of private businesses and nonprofit institutions to expand their production capacity and as an indicator of the demand for housing. Thus, movements in PFI serve as a barometer of confidence in, and support for, future economic growth.

A barometer of growth. (I said that yesterday.)

(Neither of those PDFs plays nicely with CTRL-C cut and CTRL-V paste operations.)

This post resolves two issues. First, that PFI and FPI are two names for the same thing. Second, what is measured by FRED's FPI.

Looks like a Measure of the Economy's Performance

Yesterday I concluded

So we see that when debt is relatively low, economic growth (or in this case, Fixed Private Investment) is relatively high. But when debt is relatively high, investment doesn't run higher, and sometimes it runs low.

So the next thing I want to see is Fixed Private Investment relative to Total Debt:

Graph #1: Fixed Private Investment relative to Total Debt

It looks like a measure of the economy's performance for the last 60 years.

It is.

Fixed Private Investment

Random Eyes showed me Fixed Private Investment relative to GDP:

Graph #1: Fixed Private Investment relative to GDP

Low in the early years, then three humps in the middle years, then a significant decline corresponding to the decline in total debt growth, 1986-1992. Then a couple high points that I don't recognize. But the pattern reminded me of total debt growth. So I compared the growth of Fixed Private Investment to the growth of total debt:

Graph #2: Growth Rates of Fixed Private Investment (blue) and Total Debt (red)

Low together in the early years... Rising together... Humping together... And declining together. There even seems to be similarity in the years after 1990, though the red line rides a little higher on the blue.

Definitely noticeable in the hump years, 1970-1985, the blue peaks lead and the red peaks lag. Investment seems to spike up, dragging debt along behind it. But maybe that lag was related more to inflation than to investment, for in the years before 1970 the lag is less obvious. Or maybe the debt humps were just smaller, in the early years.

I can use a little multiplication to make the up-and-down variations in the red line bigger... And then a little subtraction to bring the whole red line down and position it atop the blue line again:

Graph #3: The Red Line from Graph #2 (Debt) Scaled Up and Shifted Down for Comparison

No finesse was involved; "3" and "20" were the first numbers I tried. But multiplying to scale things up, and subtracting to line things up, are the same techniques used by Lars Christensen for his market index calculation, which I looked at here.

In the early years on Graph #3 debt (the red line) runs low relative to Fixed Private Investment (blue). They run neck-and-neck through the first two humps but in the third hump investment peters out early while debt continues to increase to the mid-1980s.

When debt growth reaches a low in the early 1990s, investment rockets up and stays relatively high for most of the decade.

In the late 1990s debt (blue) spiked up to meet investment, then fell, then investment fell. In the 2000s again debt growth rose to meet investment -- and both collapsed.


I realize that rates and levels are not the same thing. A low level of debt and a high level of debt, both growing at the same rate, may have significantly different effects on the economy. However, a low level of debt becomes a high level sooner at a high rate of growth. So rates and levels are related. We've been looking at growth rates in the graphs today, and now I will conclude by talking about levels. They're related.

What I think is, creating debt creates new money which lets spending expand; this is necessary for growth, at least under existing policy.

But creating debt also adds to the total accumulation of debt and increases the cost associated with that accumulation. This cost can hinder growth.

So we see that when debt is relatively low, economic growth (or in this case, Fixed Private Investment) is relatively high. But when debt is relatively high, investment doesn't run higher, and sometimes it runs low.


Related posts:

• Gimmie Some Slack (4 Feb 2010)
• Credit Efficiency (21 March 2010)