On Feb 21, 2009, at 5:55 AM, SA wrote:
> As for whether capital was shifted out of production, the proof is
> in the pudding. Here are the numbers I got for private fixed non-
> residential investment as a share of GDP:
>
>
> Inv/GDP
> 1930s 6.5%
> 1940s 6.7%
> 1950s 9.4%
> 1960s 9.8%
> 1970s 11.1%
> 1980s 12.1%
> 1990s 10.9%
> 2000-2007 10.7%
First off, a decline from 12.1% in the 1980s to 10.7% in the 2000s isn't trivial - it's about 12%. But these averages mask some interesting changes over time, and also by type. These numbers include equipment & software as well as buildings; according to the classic Summers-DeLong paper, equipment & software (E&S) investment is what drives productivity and growth. And they also mix financial and nonfinancial investment. If there was a boom in outfitting trading floors, this could look like a burst in "real" investment, when it's something else.
So, let's start in the 1970s, when the previous sickness of capitalism came to a head. E&S investment rose as a share of GDP throughout the decade, despite declining profitability - from 6.3% of GDP in 1971 to 8.4% in 1979. This is the sort of thing that sent Jensen et al around the bend - managers were wasting all this money on unproductive investment (from their POV, not Shane Mage's). The E&S share fell in the Volcker recession of the early 1980s, recovering with the economy, but peaking out at 8.1% in 1984 - below the 1979 peak, despite what was regarded at the time as a boom, and the recovery in profitability of the early Reagan years. It then fell as the expansion matured, to 7.1% in 1990, as the economy was about to head south, bottoming at 6.2% in 1992. It rose through most of the 1990s, along with the tech bubble (confirming SA's point that bubbles often come along with an increase in real investment - in fact, in Shiller's model, they're often inspired by new technologies). E&S peaked at 9.4% of GDP in mid-2000, as the NASDAQ was starting to leak air. It then fell through the recession and into the first years of the recovery/expansion, bottoming at 7.2% in the beginning of 2004. But it only rose modestly to 7.6% in 2005-6, and started falling even before the economy peaked. It was 6.6% in the fourth quarter of 2008. The low levels of investment came despite a strong recovery in profitability. The only reason for that that makes any sense - and Robert Gordon agreed in his papers on productivity from a few years ago - was Wall Street pressure to keep profits high in a mediocre expansion.
And that's for the entire business sector. Narrowing the focus to nonfinancial corps, whose capital expenditures aren't in the NIPAs but are in the flow of funds, and you get a much more dramatic decline - from a peak of 10.7% of GDP in 1981 to a low of 6.4% in 1992. They rose in the 1990s along with the E&S figures, but then fell more sharply, almost matching the 1992 low in 2003-4.
So putting it all together, I'd say the pressure from the financial side to limit investment was felt most dramatically in the 1980s and early 2000s. In the boom of the 1990s, everyone got carried away and had a lot more fun.
Doug