Digital Abundance and Scarce Genius: Implications for Wages, Interest Rates, and Growth

Please send this one today. Ed right is your wheelhouse, they find that "genius' share" (so talents, they define it as the top 3% of all US labor income * by the total US non-financial corporate labor income) ) of income increased from a nadir of 6.7 percent in 1984 to a peak of 31.4 percent in 2006.

"...An increasing share of income is being paid to neither traditional capital nor traditional labor. At the same time interest rates, investment rates, and total factor productivity growth are low. Informed by the economics of digitization, we provide a simple macroeconomic model that generates these relationships.... We show that when increasingly digitized capital and labor are sufficiently complementary to inelastically supplied genius, innovation in either of the first two factors can decrease wages and interest rates in the short and long run. Growth is increasingly constrained by the scarce input, not labor or capital.....Figure 5 presents the share of US non- financial corporate gross value added that is not paid to traditional labor, traditional capital, or to the government. The rst component of genius corresponds to exceptional labor income. Brynjolfsson and Saint-Jacques (2015) conceptualizes the US labor force as consisting of individuals of normal and exceptional abilities (or luck). Normal workers face a log- normal distribution of labor income which is consistent with a multiplicative model of normally-distributed abilities. But exceptional workers sort themselves into sectors with superstar-style returns, such as digital and networked industries, and draw their incomes from a Pareto distribution. This approach explains the fractal nature of top labor income percentiles. In 2008, the top 25 percent of US workers earned more than half of all labor income, the top 2.5 percent of workers earned over 20 percent of labor income, the top.25 percent of workers earned over 7 percent of total labor earnings, and so on. Using a maximum likelihood approach, they study US data and find that approximately 3 percent of workers draw from this `power-law economy' while the rest draw their incomes from log-normal distribution. They also fi nd that there has been significantly increasing skewness in this part of the economy, and that has further the increased share of income accruing to superstar workers. Accordingly, we use their 3 percent cutoff to motivate our distinction between traditional and `genius' labor. The fi nal component of genius is what is measured as pro ts: i.e. income earned by fi rms surpassing what they implicitly pay to labor, owner-operators, traditional capital, and the government. This share was slightly negative in the mid-1980s, but was 10.8 percent of corporate gross value added in 2012. While some of this share may correspond to `true' pro ts, we argue in the micro-foundations section that the majority of this share is better interpreted as payments to extremely talented individuals or returns to inelastically supplied intangible assets. All together, genius' share of income increases from a nadir of 6.7 percent in 1984 to a peak of 31.4 percent in 2006.... Roughly half of this increase is due to an increase in corporate intangible income and half from an increase in genius labor income. These trends are robust to variations in the calculation of capital's share of income (Barkai (2016) and (Barkai and Benzell, 2018)) or the precise labor income percentiles considered exceptional (Brynjolfsson and Saint-Jacques, 2015). To those in the business world, the result that non-traditional factors are increasingly scarce will come as no surprise. Many employers complain about the rarity of exceptional talent. In the words of Elbert Hubbard, "One machine can do the work of fifty ordinary men. No machine can do the work of one extraordinary man."...The good news is that when inputs can be digitized, perfect copies can be made at virtually zero cost. The bad news is that not all types of inputs can be digitized. Digital abundance leads to bottlenecks whenever an input which cannot be digitized is an essential complement. Digitization can create substitutes for many types of ordinary labor and capital, and driving down their compensation. At the same time, others earn extraordinary returns because their contributions, whether due to genius or luck, cannot be easily digitized. The most popular alternative explanation of the decrease in the traditional capital and labor share of income is increased profits. This could either be due to a decrease in oligopolistic competition or from the most pro table firms lowering their markups (slightly) while capturing a larger share of the market (Barkai (2016), De Loecker and Eeckhout (2017), Autor et al. (2017)). We see Autor et al. (2017) as a model of increasing returns to intangible assets, and therefore complementary to our paper. When industries become more competitive, there is an increased return to firms with a good productivity draw. The difference between profits and returns to unmeasured intangible assets may be a semantic one. Many have the intuition that intangible assets and superstar workers are more abundant than ever. Perhaps the most surprising thing then about our result is that these factors are increasingly scarce. We contend that this is due to confusion between the value and importance of these inputs, which are increasing, and their relative abundance, which is decreasing.We suggest several microfoundations of this aggregate relationship and explore implications. Our `microfoundations' are not mutually exclusive and may ultimately be revealed as a simpliffed representation of a complex underlying trend. But the relationship between high non-capital and labor shares, inequality, low interest rates, digital abundance and low TFP growth is a real one, and one parsimoniously captured in our framework. Perhaps, over time, Le Chatelier's principle will win out, and the bottlenecks min innovation will be overcome, simultaneously raising wages, interest rates, productivity growth and lowering inequality and genius' share. Whether or not it does, we expect these desideratum to be connected well into the future.

Seth G. Benzell and Erik Brynjolfsson, "Digital Abundance and Scarce Genius: Implications for Wages, Interest Rates, and Growth," National Bureau of Economic Research, February 2019, https://www.nber.org/papers/w25585