Investment in Equipment and Software

Image result for software graphic

Investment in Equipment and Software

Karl Smith claims that what most people think of as “Investment” is investment in equipment and software, and that it has performed well during the current modest recovery. Real investment in equipment and software has reached its previous peak. The broader measures of investment remain below their previous peaks, but this is due to residential investment and investment in structures. In Smith’s view, the problem is construction.

To some degree, Smith is seeking to respond to those “free market” economists, centrally Robert Higgs, who point out that real consumption expenditures has reached its previous peak, so that consumption isn’t a problem. Smith has also shown that real retail sales remain below its previous peak, and that the recovery in real consumption expenditures has largely been in implicit rental payments by homeowners, health care paid by Medicare and Medicaid, and the difference between the interest rates banks earn and the very low interest paid on deposits.

My view is that comparing any element of real expenditure to a peak four years ago is really beside the point. As a rule of thumb, each element of real expenditure should by now be about 12 percent higher than its previous peak. (Yes, my rule of thumb doesn’t take into account compounding.) More importantly, given that nominal GDP is 14 percent below its growth path of the Great Moderation, my first question is always to compare the nominal value of some type of expenditure to its growth path of the Great Moderation. For example, nominal consumption is nearly 15 percent below its growth path of the Great Moderation. (Real consumption is nearly 12 percent below its growth path of the Great Moderation.)

What about investment in equipment and software? It’s current value is $1,137 billion, which has just surpassed its peak of the third quarter of 2007. The trend growth rate of investment in equipment and software during the Great Moderation was 5.92 percent. (From quarter 1 1985 to quarter 4 2007.) This is slightly higher than the trend growth rate of nominal GDP, which was 5.4 percent.

And so what would investment in equipment and software be if it had continued to grow at trend? It’s current value would be $1,517 billion. It is currently 24 percent below its trend growth path from the Great Moderation.

Looking at the diagram, perhaps the trend is pushed up by the huge increase right before the 2001 recession. (On the other hand, it is possible that the “problem” was that investment in equipment and software was crowded out by investment in housing.) Interestingly, only a slightly reduced growth rate, to 5.7 percent, puts actual investment very close to this modified trend before the Great Recession.

Even with this slightly slower growth path for investment in equipment and software, it currently remains 17.8 percent below trend. (And if it had been growing with the rest of nominal GDP at 5.4 percent, its current value would be 10% below trend.)
And what about real investment in equipment and software?

Real investment in equipment and software is 28 percent below the trend of the Great Moderation! Like the nominal series, it just passed its previous peak. And while there have been some very high quarterly growth rates during the recovery, (21 percent in the second quarter of 2010,) the collapse was steep during the recession (37 percent in the first quarter of 2009.)
What about the price index for equipment and software? The trend for the Great Moderation was 1.2 percent deflation. And while the price index is currently 5 percent above trend, it has been remarkably stable since 2002 (at 100!) Remarkably, something like the modified nominal series, showing investment 18 percent below trend is likely much more “real” than deflated (that is inflated,) series.

In my view, monetary disequilibrium has depressed nominal expenditure on output across the board. Both consumption and investment in equipment and software have been depressed. That the real value of some element has finally passed its previous peak provides approximately no reason to believe that its current nominal value is appropriate.