Carol Corrado, Charles Hulten, and Daniel Sichel write,

the fraction of output growth per hour attributable to the old “bricks and mortar” forms of capital investment (labeled “other tangible” capital in the lower panel of table 5) is very small, accounting for less that 8 percent of the total growth in the period 1995-2003. While it is inappropriate to automatically attribute the other 92 percent to “knowledge capital” or “the knowledge economy,” it is equally inappropriate to ignore the association between innovation, human capital, and knowledge acquisition, on the one hand, and investments in intangibles, IT capital, labor quality change, and multifactor productivity, on the other.

Their paper looks at intangible investments made by firms. Treating intangible investment as capital in the national income accounts would raise estimated annual labor productivity growth for 1995-2003 from 2.78 percent to 3.09 percent, while reducing the amount of unexplained productivity growth (“multifactor productivity”) from 1.42 percent to 1.08 percent.In my business career at Freddie Mac and with Homefair.com, I was always working on innovation, as opposed to producing and distributing output. The issue came up in an accounting context in both firms. At Freddie Mac, I helped launch an initiative to use credit scoring to automate and streamline the underwriting process. When senior management finally embraced this idea, they chose to put together a huge project and to capitalize the expenses for that project. This was a bitter period for me (I was shunted aside and soon quit the company), so I really cannot comment objectively about it.

At Homefair, we were bought out by a large newspaper company, but the terms were an “earn-out,” where we continued to run the company and our final payment was to be a multiple of EBIDTA, which is earnings before interest, depreciation, taxes, and amortization. As the Internet boom heated up (this was 1998), it seemed as though a more aggressive growth strategy was called for. To align the incentives properly, our parent company agreed to count some of our marketing and product development expenses as capital expenses, so that we could use depreciation and amortization accounting, raising our EBIDTA. Again, I cannot say whether that was truly proper. In the end, our parent company sold us to a high-flying Internet company that went public in 1999, so the whole EBIDTA formula was moot–buried under a pile of dotcom-era cash and stock. The cash retained its value.