Macro Roundup

"Educated voters’ leftward shift is surprisingly old and international,"The Economist, May 29, 2021, https://www.economist.com/graphic-detail/2021/05/29/educated-voters-leftward-shift-is-surprisingly-old-and-international

Educated voters’ leftward shift is surprisingly old and international

Thomas Piketty, an economist, became famous for a book that analysed 200 years of data on wealth inequality in a wide range of countries. This month he published a paper, co-written by Amory Gethin and Clara Martínez-Toledano, which applies a similar approach to the relationship between demography and ideology. Its findings imply that the electoral victories of Donald Trump and the Brexit campaign in 2016 were not an abrupt departure from precedent, but rather the consequence of a 60-year-old international trend.

In a paper in 2018 Mr Piketty noted that elites in Britain, France and America were split between intellectuals who backed left-of-centre parties—he dubbed them the “Brahmin left”—and businesspeople who preferred right-wing ones (the “merchant right”). His new work expands this study from three Western democracies to 21. It combines data on parties’ policy positions with surveys that show how vote choices varied between demographic groups.

The paper finds that income and education began diverging as predictors of ideology long ago. In 1955 both the richest and the most educated voters tended to support conservative parties. Conversely, both poorer and less-educated people mostly chose labour or social-democratic ones.

Today, wealthy people still lean to the right. In contrast, the relationship between education and ideology began to reverse as early as the 1960s. Every year, the 10% of voters with the most years of schooling gravitated towards left-wing parties, while the remaining 90% slid the other way. By 2000, this had gone on for so long that, as a group, the most educated voters became more left-wing than their less-educated peers. The gap has only grown since then.

This trend is strikingly consistent. It developed just as fast in the 20th century as in the 21st, and appears in almost every Western democracy studied. This includes both two-party systems and proportional ones, in which green parties now lure educated voters, and nativist parties attract the less educated. Such breadth and regularity make the rise of right-wing populists like Mr Trump—and of left-of-centre technocrats like Emmanuel Macron or Justin Trudeau—look like a historical inevitability.

Economist on changing patterns in voting behavior across advanced economies"...Today, wealthy people still lean to the right. In contrast, the relationship between education and ideology began to reverse as early as the 1960s. Every year, the 10% of voters with the most years of schooling gravitated towards left-wing parties, while the remaining 90% slid the other way. By 2000, this had gone on for so long that, as a group, the most educated voters became more left-wing than their less-educated peers. The gap has only grown since then. This trend is strikingly consistent. It developed just as fast in the 20th century as in the 21st, and appears in almost every Western democracy studied. This includes both two-party systems and proportional ones, in which green parties now lure educated voters, and nativist parties attract the less educated. Such breadth and regularity make the rise of right-wing populists like Mr Trump—and of left-of-centre technocrats like Emmanuel Macron or Justin Trudeau—look like a historical inevitability...."

Gort and Klepper present evidence that the product show a cycle of large new entries followed by shakeouts, the basic dynamic, “….Stage I encompasses the interval in which the number of producers in the market remains relatively small (usually between one and three). Stage II is the interval from the 'take-off' point of net entry to the time that net entry decelerates drastically. Stage III is the ensuing period of low or zero net entry, and Stage IV is the subsequent period of negative net entry. Stage V represents the new equilibrium in the number of producers that coincides with the maturity of the product market and continues until some new fundamental disturbance generates a change in market structure…”

Michael Gort and Steven Klepper, "Time Paths in the Diffusion of Product Innovations,"The Economic Journal, September 1982, http://www.unife.it/economia/lm.economia/insegnamenti/economia-applicata-avanzata/mat-did/frattini/ff1-gort1982.pdf

Their takeaway, “…(1) The markets for most new products appear to pass through at least five distinguishable stages in the course of their evolution. (2) New industries generally pass through a stage in which the number of producers declines significantly.(3) The evidence does not support the hypothesis that variations in entry of producers into new markets can be explained largely by economies of scale. (4) The dynamic adjustment costs hypothesis as an explanation of entry rates is consistent with some of the evidence but is not a sufficient explanation for many of the observed phenomena. (5) There appears to be an association between rises and declines in the rate of innovation and the rate of entry into new markets. We interpret the causal relation as being positive, and flowing primarily from innovations to entry rates during the period of positive net' entry. (6) The character, importance, and sources of innovations appear to change over the product cycle. (7) The results support the conclusion that the structure of markets (in terms of number and composition of producers) is shaped, to an important degree, by discrete events such as technical change and the flow of information among existing and potential producers…”

“…Results of the decomposition are presented in Tables 2 and 3. For each product, Table 2 lists the number of years classified in each stage and Table 3 lists the mean rate of net entry of the years classified in each stage….Notwithstanding considerable variation in the duration of stages among products, certain dominant attributes of the process of entry stand out in Tables 2 and 3 with remarkable consistency. First, as Table 2 demonstrates, there are few instances of initial commercial introduction of a product that are immediately followed by rapid entry. For all but three of the forty-six innovations, there was at least one year in Stage I (preceding take-off in entry). Closer examination of Table 2 indicates that the average length of Stage I has declined over time. While the overall average length of Stage I is I4j4 years, the average length of Stage I for products introduced before I930 was 23-I years; it was 9-6 years for those introduced in I930-9 and only 4-9 years for products introduced in I940 or later. While the result could arise partly from sample selection bias, it strongly suggests that the interval required for successful imitation has systematically declined over time. Table 2 further indicates that, of the 36 products which had attained Stages IV or V by I973, all proceeded through a distinct contraction phase in the number of producers. As indicated in Table 3, the average annual rate of net entry in this period for the 36 products was - 4-84 firms. The average duration of the stage was 5-4 years - roughly half the average duration of 9.7 years for Stage II…”

“…we assembled data on the historical development of 46 new products.' The 46 were chosen on the basis of three criteria: (I) to allow sufficient diversity by including consumer, industrial, and military products; (2) to include only products that were 'basic' innovations; (3) to include products with adequate data on net entry. The 46 products are listed inTable 2 along with the year they were first commercially introduced.2 The initial dates of commercial introduction span a 73-year period, beginning with phonograph records in 1887 and ending with lasers in 1960…”

The evidence,

Klepper/Graddy argue that they“..establish that the number of firms in new industries follows a distinctive path: first it grows, then declines sharply, and finally levels off…”echoing a cycle of new entries followed by shakeouts

Their takeaway, “…Analyzing first the data on the number of firms, each product history was divided into a maximum of three stages. Stage 1 is composed of the years in which the number of firms grows. Stage 2 is characterized by a decline or shakeout in the number of firms. Stage 3 is composed of the period after the shakeout when the number of firms stabilizes…One factor bearing on the evolutionary process is chance. In each period, potential entrants randomly receive information that qualifies them for a particular cost. The realization of this random process will have an important effect on an industry’s ultimate market structure. To see this, note that the model predicts thatthe industry will even be composed of only those firms that attain the lowest possible cost after imitation, with the largest firms being those that were among the first to attain this cost. If, after the first N firms attain this cost, there is an unexpectedly long period before other firms receive the information required to attain this cost, then the N-firm concentration ratio of the industry will be high, ceteris paribus. Indeed, if there is an unexpectedly long delay after the first firm attains the lowest possible cost, then the industry could end up as a virtual monopoly…”

Steven Klepper and Elizabeth Graddy, “The Evolution of New Industries and the Determinants of Market Structure,” The RAND Journal Of Economics, Spring 1990, https://www.jstor.org/stable/2555491

Their evidence, “…Table 1 lists the 46 products and the number of years in each stage for each product. The table indicates that by 1981, all but 8 of the products had attained stage 2, and 22 of the 46 products had attained stage 3. The 8 products that had not attained stage 2 by 1981 are relatively young, with a mean birth year of 1945 versus 1926 for the other 38 products. Thus, it might be expected that eventually these products will also pass through a shakeout period before reaching maturity. While the 46 products appear to follow a similar pattern in terms of the three stages in the number of producers, Table 1 indicates that the length of each stage varies considerably across products. This is particularly true for stage 1, where some products have a much longer gestation period before developing a sizable market.9 The average number of years in stages 1 and 2 across all products attaining the next stage is reported at the bottom of Table 1. Table 2 lists the mean annual change in the number of producers in each stage for each product. Not surprisingly, in each stage there is considerable variation across products in the mean annual change in the number of producers, particularly in stages 1 and 2. What is more surprising is the severity of stage 2 (the shakeout) and the relative stability in the number of producers in stage 3. The severity of stage 2 is conveyed by Table 3, which reports the net decrease in the number of producers in stage 2 as a fraction of the peak number of producers for the 22 products attaining stage 3 by 1981. On average, Table 3 indicates there was a net decrease of 52% of the producers in stage 2, with some industries experiencing a shakeout of over 80% of their producers. In contrast, there is no clear trend in the number of producers in stage 3 for the same 22 products: 11 experience a net increase, 8 a net decrease, and 3 no change at all in the number of producers...”

New McKinsey report breaks down large firms in OECD countries three large takeaways, first their impact on GDP, “…The business sector overall contributes 72 percent of GDP in the OECD, and corporations with more than $1 billion in revenue account for an increasingly large share of that.Measured in real GDP per capita, the contribution of the business sector— defined as for-profit companies of all types including corporations, partnerships, and sole proprietorships—has tripled since 1960 on average in major OECD economies, in proportion with those economies’ overall growth. Companies underpin 85 percent of technology investment and 85 percent of labor productivity growth since 1995, a larger proportion than their GDP contribution. The size of the business sector varies only modestly within each of the major economies, and its share has remained steady for the past 60 years. However, this steadiness masks significant underlying shifts, notably the growth of corporations with more than $1 billion in revenue, which increased their global value added by 60 percent relative to their home countries’ GDP since 1995….”
Their breakdown of how value flows from large firms to households largely via labor income and consumer surplus, “…Economic value flows from companies to households via eight pathways, ofwhich labor income and consumer surplus are the largest direct pathways. Our mapping work identifies eight pathways through which economic value from corporations flows to households and the economy. Five are directly measurable monetary flows: labor income, capital income, taxes, investment in capital assets, and payments to suppliers. The money flowing through the supplier and investment pathways passes through other companies to reach households and the economy. The sixth is consumer surplus (the difference between what consumers are prepared to pay and what they do pay), which we estimate. The final two pathways are negative and positive spillovers, which we do not assess comprehensively beyond the examples of environmental impact and contributions to total factor productivity growth (unaccounted for in additional labor or capital input). Based on our data set of about 5,000 corporations with more than $1 billion in revenue,we find that labor income is the largest direct measured pathway, with wages and benefits accounting for $0.25 flowing to households for each dollar of revenue. Just over half of every dollar, $0.58, goes to suppliers (themselves companies large and small), reflecting the role suppliers play in enabling corporations to create and deliver their products and services. The other significant pathway is consumer surplus, which we estimate to be about $0.40 per dollar of revenue….”
How the pathways have changed overtime, “…The size of the pathways has changed over the past 25 years, with capital income growing while labor income and supplier payments have declined. Between the periods 1994-96 and 2016-18, the biggest change has been a two-thirds increase in the capital income pathway, from $0.04 to $0.07 per revenue dollar. Applied to the $40 trillion in revenue represented by our large company data set, this $0.03 difference amounts to an increase in capital income of over $1 trillion. The labor income pathway shrank by $0.02, or 6 percent, as labor per dollar of revenue fell by 15 percent and wages grew only 11 percent. Productivity gains amounted to 25 percent in real terms, significantly more than wage growth, as gains from labor productivity went predominantly to capital income. Investment in intangible assets tripled as a proportion of revenue in this period, while investment in tangible assets dropped by half. Supplier payments fell by $0.02, or 4 percent. The decline was especially steep for suppliers that were small and midsize enterprises, which saw a 10 percent drop in share of payments to suppliers in the United States, with similar patterns in other countries. The share going to domestic suppliers in each country also fell; about half of the decrease in supplier payments moved to foreign OECD suppliers and the other half to non-OECD suppliers. The tax pathway held steady overall as decreases in corporate taxes, especially in the United Kingdom and the United States, were offset by slight increases in production taxes, especially in Japan, and in corporate income tax in France….Decline in the share of the labor income pathway. Labor income from the corporations with annual revenue above $1 billion grew from just under $4 trillion to almost $10 trillion over the past 25 years (in 2018 dollars). However, this represents a lower share of the value that these companies add to the economy, as the labor portion of household income attributable to these corporations declined from 67 percent to 59 percent of the total.That difference is the equivalent of $1.2 trillion. This represents a steeper overall decline than domestic economies in the OECD generally. Japan saw the biggest decrease at more than ten percentage points, followed by Germany (five), the United States (four), and France (three). The United Kingdom saw a rise of four percentage points. A declining labor share puts pressure on incomes of lower-income households, who rely predominantly on income from wages for their livelihoods…. 5. Two diverging trends in the consumer surplus. On one hand, consumers in all household income segments have benefited substantially from the consumer surplus generated by large corporations, especially that from Technologists and Makers. The gains to consumers in OECD countries from these two company archetypes in our data set over the past 25 years amount to about $8 trillion in consumer surplus, compared with a scenario under which their prices of products and services stayed constant. On the other hand, price increases by Experts and Discoverers in our data set, especially for health and education services, correspond to more than $2 trillion in lost consumer surplus. This puts some of these essential services out of the reach of low-income households, at a time when the need for them has risen…”

McKinsey’s description

Approach 2: (Require time series variation in prices)

1. Assume price decrease is pure gain for consumers + additional gains because of more sales e.g. price falls from 5 to 4; assume consumers capture that whole dollar of value plus assume expanded sales because the price is now lower.
2. Use “hedonic price deflators” to adjust products for quality changes (really feature changes).
3. Bottom line - Consumer Surplus = (-1) * change in price * total quantity at new prices * quality adjustment …”

Ben Comment on McKinsey’s consumer surplus calculation: “A little bit of a black box but we can make some headway: Approach 1:

1. Assume firms and consumers split value equally
2. Look at firm's/industry’s value added and assume that this is the firm’s/industry’s portion of value (so consumer gets the same amount)
3. Adjust value added (therefore consumer surplus) “based on evidence of markups above factor costs.” Factor costs are estimated as the WACC (within firms or across industries).
4. Bottom line - looks like Consumer Surplus = price - WACC

So FRED uses the Penn World Tables For theirLabor Share,in 2019 US was.597 (in GDP at Current National Prices), in 2007 it stood at.604, down from.64 in 2001. McKinsey’s labor share number mirror that (This is from a2019McKinsey Note), not that 57% is likely lower than the 60% in the current report as the later looks only at large firms that are presumably more efficient than the broader economy.