Until a few years back, only expenditures on physical capital (business plant, equipment and vehicles, and the change in business inventories) and residential construction were included in the definition of investment. Then, business expenditures on software were added. Now, expenditures on R&D are included.
As a result of today’s change in the definition of R&D, GDP is considerably higher; but, because past estimates of GDP have all be re-stated in accordance with this definition, percentage changes in GDP are hardly affected.
Among others, I have contributed to the literature on the recognition of expenditures on R&D as investment (my work concerned stock market valuation of R&D intensive firms, and the behavior of corporate investment re-defined so as include R&D); and, so, I am particularly happy to see this.
At the time the former definition of GDP was adopted, R&D was much less significant than what it is today (although even then, it wasn’t nothing). In the US, non-defense R&D amounts to about half of business investment in physical capital; and, in Germany and Japan, it approximates business investment in physical capital. So, the non-recognition of R&D had come to distort our figure for our net national saving, among other figures, meaning, that we weren’t as pathetic savers as we appeared to be.
In my opinion, there are two other major problems with the definition of GDP. The first concerns work performed within the household (what we economists would call an imputation). As women entered the market economy, this work diminished. The GDP reflects, e.g., the purchase of value-added products such as butter, but did not previously reflect the value of churning butter within the household. While entry into the market allows more specialization and greater economies of scale in production and, so, increases GDP; the non-recognition of work performed in the home overstated the increase in GDP when women entered the market economy. Even today, the non-recognition overstates the difference in GDP between traditional economies and highly advanced economies.
The second continuing major problem in the definition of GDP concerns higher education or human capital more broadly speaking. The analogy of human capital to R&D should be obvious. They are both forms of “intangible capital.” Arbitrarily distinguishing expenditures on higher education and vocational education and on employee training by business from primary and secondary education, we are left with an amount comparable to expenditures on R&D.
Like R&D, at the time of the adoption of the former definition of GDP, expenditures on human capital were much smaller; and, today, each is quite large. The economy, itself, has been transformed from a physical capital-based, manufacturing-oriented economy, to an information-based, service-oriented economy. Companies manage a process of continuous improvement; and, individuals must commit themselves to being life-long learners.
To be sure, there are problems concerned with the recognition of expenditures on human capital as investment, even beyond the problems associated with the recognition of R&D as investment. Number one, a big part of investment in human capital is implicit in the earnings foregone while a person is engaged in a program of learning or is mastering a skill. Number two, as with public infrastructure and heavily-subsidized private investment, we wonder how sharp decision-makers are with respect to higher education expenditures in terms of the expected risk-adjusted rate of return.
Nevertheless, what is clear is that, even with the change related to R&D, we are still not fully recognizing investments in intangible capital. Were we to also recognize what is reasonable to recognize of expenditures on human capital, we would have another substantial increase to GDP and to net national savings.