From July 2013, the U.S. Bureau of Economic Analysis (BEA) will alter the way it tabulates the United States’ GDP figures (see the BEA’s Powerpoint-style cheat sheet on the revision). In a bid to “capture the shape of modern economies” (The Economist) and have GDP reflect business investments in inventions, intellectual property and copyrights, the BEA has added into the pre-existing mix of aggregate output or GDP variables (the C + I + G + X – IM equation of economics textbook lore) several “intangible” expenditure items that were previously either largely ignored in the official GDP calculus . Three such intangibles, research and development (R&D) spending, royalty receipts from “artistic originals” such as film, literature and music, and accruals from pension plans (“pension accounting”), are now counted as part of fixed investment, or “anything that can be used repeatedly or continuously in the production process for more than one year” (FT). Any good economics textbook will of course tell you that fixed investment is simply a synonym for Investment, or the (I) component of GDP. The scale and implications of this tweak in methodology are unprecedented – the BEA has revised U.S. GDP figures going all the way back to 1929 to account for the categorical changes. Despite the BEA’s practice of revising its methodology for calculating U.S. GDP every five years (this year’s is, to give its full flourish, its 14th Comprehensive Revision of the National Income and Product Accounts), this year’s tweak is arguably one of the most profound changes to U.S. national income accounting since the system’s institution in 1947: so much so that the FT humorously calls the exercise a “Hollywood makeover” for the U.S. economy, and has put up a bunch of charts that put the revision in graphical perspective.
And what an apt tag that is. The expected net result of the new methodology is a 3% jump in expected 2013 GDP. Considering fresh U.S. macroeconomic data showing a 12% improvement in domestic house prices (according to May’s reading of the S&P/Case-Shiller Index of 20 U.S. cities), a second quarter economic growth of 1.7% (up from 1.1% in the first quarter, the addition of a less-than-enrapturing but no less encouraging 162,000 jobs in July (shy of the predicted 185,000), the jump in manufacturing sentiment as recorded by a PMI rise to 56 from 52 (beating expectations of a hop to 53) and the dip in the unemployment rate from 7.6% to 7.4% – all culminating in a welcome drop in U.S. bond yields in July (but ho-ho, all eyes on the debt ceiling circus that kicks off in September) -, an upward bump for overall growth figures will only further boost consumer sentiment and keep the steady U.S. recovery well on track. Yes, it would also reignite that festering quantitative easing tapering debate, but that’s something for another post.
The retrospective revision going back to 1929 would however have a minimal impact on historical U.S. GDP figures, as the BEA graph covering the period 1970~2011 suggests.
Historically and according to received macroeconomic wisdom, expenditure items other than private consumption of goods and services, fixed investment (limited to so-called ‘residential investment’ and ‘non-residential investment’), government expenditure on goods and services and net exports have been excluded from GDP statistics, or at the very least, factored into any of these main categories as costs related to but not independently constitutive of the production process. Under the new GDP methodology, the most significant new item, R&D spending, will no longer be treated as an intermediate input or expense, but will instead be categorized as a capital investment asset under the broader fixed investment rubric. Theoretically, this would mean that spending on things such as inventions, patents and trade marks would directly add to the aggregate value of the economy. This is good news for the aggregate economy: R&D accounted for around 3% of U.S. GDP in 2009 (the figure was slightly down to 2.6% in 2012), while the reconceptualization of R&D spending is expected to boost the profile of R&D in general and spur a host of facilitatory measures and inducements such as tax rebates and subsidies for innovators.
The inclusion of R&D investment in the GDP calculus is a commonsense one. As early as 1932, the British economist John Hicks suggested in his The Theory of Wages that spending on innovations and technological improvements can have a positive effect on broader economic performance in three ways. First, where any given investment of capital in labor results in the marginal product of that capital increasing more than the marginal product of labor, any technical progress achieved is said to be “labor-saving”, increasing the demand for that particular type of capital – this implies a rise in Investment (I). Second, where capital investment in innovations results in technological enhancements that reduce costs and expenditures in other areas of the economy (for instance, investment in computer enhancements that shrink office expenditures), that capital becomes more productive and has the potential positive ripple effect of raising labor productivity as workers accomplish tasks more quickly and efficiently. And third, where capital investment in labor results in the marginal product of labor increasing more than that of capital, this increases the demand for that specific type of labor. As the price of capital falls relative to the price of labor, more people would be motivated to invest capital in allied innovations that make labor more productive. In each of these scenarios, Hicks’ takeaway message appears to be that technological change as a result of innovation spending does not have a neutral effect – rather than increasing the marginal productivity of all factors of production simultaneously, innovation spending augments particular innovations more than others, in a way that makes technological change “biased”.
How would other economies fare against the United States if they too factored spending items such as R&D expenditure into their GDP calculi?