Edward Conard

Top Ten New York Times Bestselling Author

  • “Unintended Consequences is full of substance, it is one of the must-read books of the year, and once I finish it I will be giving it a second read through right away.” - Tyler Cowen, Professor, George Mason University
  • “…reminds us that inequality sends a signal of what society lacks most, in America’s case, entrepreneurship and risk taking.” - Lawrence Lindsey, CEO, The Lindsey Group, former Director of the National Economic Council
  • “Unintended Consequences offers deep and well-argued analyses on almost every issue.” - The New York Times
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…a very valuable contribution.” - Larry Summers, former Secretary of the Treasury and director of the National Economic Council, president emeritus, Harvard University
  • “…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
  • “…challenges misconceptions that distort our economic debates.” - Arthur Brooks, President of the American Enterprise Institute
  • “…challenges misconceptions that distort our economic debates.” - Arthur Brooks, President of the American Enterprise Institute
  • “…a fresh argument for the productive value of inequality.” - David Autor, Professor of Economics, Massachusetts Institute of Technology
  • “…a very valuable contribution.” - Larry Summers, former Secretary of the Treasury and director of the National Economic Council, president emeritus, Harvard University
  • “…a must-read for serious students of economic policy.” - Glenn Hubbard, Dean, Columbia Business School, and former Chairman of the Council of Economic Advisers
  • “Unintended Consequences provides a provocative interpretation of the causes of the global financial crisis and the policies needed to return to rapid growth. Whether you agree or not, this analysis is well worth reading.” - Nouriel Roubini, New York University; Chairman, Roubini Global Economics
Upside of Inequality Oxford Unintended Consequences

Macro Roundup


Monday, May 17, 2021

AI Summary. U.S. manufacturing market concentration remained stable once foreign firms' sales are included in the measures, with foreign firms' market shares doubling from 6.9% to 14.4% among lower-ranked firms while staying below 3% for top-ranked firms. The authors estimate import competition accounts for half of the decline in concentration among the top 20 U.S. firms.

Mary Amiti and Sebastian Heise Federal Reserve Bank of New York
Date Posted:
Is Database:
Database

US manufacturing market concentration remained stable btw 1992-2012 when accounting for foreign firms’ sales.

Does including foreign firms' sales change measured U.S. manufacturing concentration?

Core argument: U.S. manufacturing market concentration remained stable between 1992 and 2012 once foreign firms' sales were included, with the inclusion of foreign firms mitigating measured concentration increases across industries.

Between 1992 and 2012, U.S. manufacturing market concentration remained stable when accounting for foreign firms' sales. The inclusion of foreign firms in market concentration measures mitigated the rise in concentration across various industries. Foreign firms increased their presence, particularly in the lower sales distribution, counteracting the concentration growth among U.S.-based firms. Industries with high import penetration in 1992 saw a decline in concentration due to increased import competition, while those with low import penetration experienced a rise. Import competition led to a reduction in the market share of large U.S. firms, with foreign firms' market shares doubling from 6.9% to 14.4% in lower ranks, while remaining below 3% for top-ranked firms. This dynamic suggests that import competition accounts for half of the decline in concentration among the top 20 U.S. firms, highlighting the importance of including foreign sales in market analysis.

Takeaways by Macro Roundup® AI

  1. U.S. manufacturing market concentration remained stable between 1992 and 2012 once foreign firms' sales were included, with the inclusion of foreign firms mitigating measured concentration increases across industries.
  2. Foreign firms' market shares in lower-ranked positions doubled from 6.9% to 14.4% while remaining below 3% among top-ranked firms, counteracting concentration growth among U.S.-based firms.
  3. Industries with high import penetration in 1992 saw concentration decline while those with low import penetration saw it rise, and import competition appears to account for half of the concentration decline among the top 20 U.S. firms.

Core findings, one"…Market concentration in U.S. manufacturing, adjusted for foreign firms, remained stable between 1992 and 2012. ... once the market shares of foreign firms are taken into account, we find that market concentration did not rise but instead remained flat between 1992 and 2012.... Furthermore, our results are not driven by just a handful of industries. We find that the inclusion of foreign firms in the market concentration measure attenuates the rise in concentration between 1992 and 2012 in a broad range of industries...."

Two, "...Foreign firms have increased their presence among the top 20 firms, but their share in the top 20 remains low. Foreign firms’ largest growth has been in the bottom part of the sales distribution…we reconcile the flat trend in overall market concentration with the previously documented rise in domestic concentration by showing that the growth of foreign firms’ market shares was mostly at the bottom of the sales distribution...Our comprehensive firm-level data allows us to pinpoint which part of the market share distribution foreign firms enter. We find that the entry of foreign firms with small market shares counteracted the increase in concentration among U.S.-based firms, generating the flat trend in overall concentration..."
Three, "…. market concentration fell mostly in those industries with high import penetration in 1992, which are also those industries that experienced the fastest growth in import competition over the following 20 years. In contrast, concentration rose in industries with low import penetration. Effectively, domestic and overall concentration differ the most in industries where foreign firms play a significant role. For example, import-competing industries, such as electronics, are the ones that were further liberalized and became less concentrated. By contrast, industries like concrete remain fairly closed to trade and thus do not face the increased competition from foreign firms...."
Evidence, “...Is the stable trend in market concentration driven by a few large industries or does foreign competition reduce concentration more broadly? To explore this question, Figure 2 plots the change between 1992 and 2012 in the top 20 domestic concentration measure on the x-axis against the change in the top 20 overall market concentration measure on the y-axis as a bin scatter. We bin the industries by ranking them by the change in their domestic concentration measure, and then combine them into 20 groups of 8-9 industries each.7 Each bubble depicts one of these groups, with the size of the bubble proportional to the industry group’s total absorption, defined as total shipments by U.S. firms minus exports plus total imports. The figure shows that nearly all of the bubbles are below the 45- degree line, indicating that accounting for foreign firms’ sales in the U.S. results in a smaller increase in concentration in almost all industry groups than the domestic concentration measures would suggest. However, there is a wide range in the size and direction of changes in concentration, and even with the full measure concentration rose in a number of industries, as shown by the bubbles in the top right quadrant. The reason the entry of foreign firms did not increase market concentration is because their entry and growth was mostly in the bottom part of the market share distribution, with the loss in domestic shares completely offset by the gain in foreign shares as predicted by models with a Pareto distribution. Combining the foreign firm-level sales data with those of U.S. firms reveals the following novel fact…. Thus, the height of the first bar shows that the market share of all of the number 1 ranked firms amounts to 12 percent of all manufacturing sales in 1992.8 We split each bar into the market share accounted for by domestic firms in blue and by foreign firms in red. It turns out that foreign firms with rank 1 account for a market share of virtually zero in the aggregate. This is mostly due to the fact that there are very few industries where foreign firms rank in the top 20. Consequently, nearly all of the market share of firms with rank 1 is accounted for by domestic firms. To see how these patterns evolved over the sample period, we plot the analogous information for 2012 with lighter colors. A clear pattern emerges, showing that the largest growth in foreign sales is in the bottom part of the distribution. Foreign firms with a rank higher than 50 more than doubled their market share from 6.9 percent to 14.4 percent. By contrast, the foreign shares of the top ranked firms remained low on average, below 3 percent for each of the top seven bins..."

U.S. Market Concentration and Import Competition: Extended Excerpt Image 1

U.S. Market Concentration and Import Competition: Extended Excerpt Image 2


Bottom line, "...We show that once foreign firms’ sales in the U.S. are taken into account, market concentration did not rise but instead remained flat between 1992 and 2012. We reconcile the flat trend in market concentration with the previously documented rise in domestic concentration by showing that the growth of foreign firms’ market shares was mostly at the bottom of the sales distribution, counteracting the increase in concentration among U.S.-based firms. Consistent with the theory, we show that import competition caused an increase in domestic market concentration among U.S. firms as well as exit. In contrast, import competition caused the largest U.S. firms to lose sales as a share of total sales in the U.S. market, inclusive of both foreign and domestic firms...our results suggest that import competition accounts for half of the overall decline in market concentration of the top 20 U.S. firms. These contrasting results highlight the importance of the definition of a market in understanding the relationship between import competition and concentration. When we only consider the sales of the top U.S. firms relative to the sales of other U.S. firms, we find that larger firms gained market share in industries with tougher import competition, and hence concentration appears to have risen. However, comparing the market share of the top U.S. firms to all sales in the U.S. market, inclusive of foreign firms’ sales, we find that the sales share of large U.S. firms actually shrank….”
Their findings have implications for views on market power, "... Interpreting our findings through the lens of these models suggests that markups of domestic firms have fallen and those of foreign firms have risen, offsetting each other resulting in stable aggregate markups...."
Mary Amiti and Sebastian Heise, "U.S. Market Concentration and Import Competition," Federal Reserve Bank Of New York, May 2021, https://www.newyorkfed.org/medialibrary/media/research/staff_reports/sr968.pdf

Ben Comment:“…It’s hard to get a sense of exactly how well their import data maps into the CMF. They say they don’t count US-firm imports because that’d be double counting. Im not sure that that’s right. The CMF is an establishment level data set and if the establishments are overseas, I don’t know how their sales would get into it. If they’re not counting US company foreign production, that would be a big issue since concentration is exactly the thing they’re trying to measure. I’m not surprised their discussion here is sparse. There are big disclosure issues at the census so it’s not really surprising. They also have a model and a few regressions but those are all secondary to the headline result - the whole concentration literature, at least in manufacturing, is wrong. Certainly throws a wrench in Eeckhout’s argument (in manufacturing)….”

Research from FRBNY using census data from the American manufacturing sector finds"..we show that typical measures of concentration, once adjusted for sales by foreign exporters, actually stayed constant between 1992 and 2012..Although concentration among U.S.-based firms rose, the growth of foreign firms, mostly at the bottom of the sales distribution, counteracted this increase. We find that higher import competition caused a decline in the market shares of the top twenty U.S. firms..."

Wednesday, May 19, 2021

AI Summary. Among top earners, a 1% rise in the national unemployment rate is associated with a 0.5% lower likelihood of employment at elite consulting, finance, and technology firms and a 0.2% higher likelihood of entrepreneurship, about 15% above the 2.2% baseline. Startups formed under these conditions outperform average startups.

Isaac Hacamo Social Science Research Network
Date Posted:
Is Database:
Database

Top earners are more likely to become entrepreneurs when labor market opportunities decline in top consulting, finance, and technology firms.

Does weak hiring at elite firms push top earners into entrepreneurship?

Core argument: A 1% rise in the national unemployment rate is associated with a 0.5% decline in the likelihood of employment at top consulting, finance, and technology firms and a 0.2% increase in the likelihood of entrepreneurship among top earners.

Economic data reveals that top earners are more likely to become entrepreneurs when labor market opportunities decline in top consulting, finance, and technology firms. A 1% increase in the national unemployment rate correlates with a 0.5% decrease in employment likelihood at these firms, leading to a 0.2% increase in entrepreneurship likelihood. This shift is significant given a baseline entrepreneurship rate of 2.2% at graduation, translating to a 15% increase relative to the mean. Notably, startups formed under these conditions often outperform average startups, with 15% receiving venture capital, 4.5% being acquired, and 6% obtaining patents. The majority of these firms emerge in financial, professional, and technical services sectors, reflecting the skillsets of their founders. This trend underscores the potential for labor shocks to enhance entrepreneurial quality and innovation.

Takeaways by Macro Roundup® AI

  1. A 1% rise in the national unemployment rate is associated with a 0.5% decline in the likelihood of employment at top consulting, finance, and technology firms and a 0.2% increase in the likelihood of entrepreneurship among top earners.
  2. Against a baseline entrepreneurship rate of 2.2% at graduation, the 0.2% increase tied to weaker labor market opportunities amounts to a 15% rise relative to the mean.
  3. Startups formed under these labor market conditions often outperform average startups, with 15% receiving venture capital, 4.5% being acquired and 6% obtaining patents, and most emerge in financial, professional, and technical services, reflecting their founders' skillsets.

Forced Entrepreneurs: Extended Excerpt Image 1


Isaac Hacamo and Kristoph Kleiner, "Forced Entrepreneurs," Social Science Research Network, May 2021, https://papers.ssrn.com/sol3/papers.cfm

They define the most selective schools as, “…we define the ten most selective colleges as University of Chicago, Harvard University, Columbia University, Stanford University, Princeton, CalTech, UC Berkeley, MIT, Yale University, and University of Pennsylvania…”

“…estimating that (i) 15% of Fortune 500 firms are created by alumni of top 20 collegesand (ii) these firms represent 30% of Fortune 500 employment…. Our findings are additionally supported by evidence from Black (2020) that over 20% of all venture capital funding is given to graduates of top 20 universities…. Finally, Bhagat et al. (2010) estimate over 15% of all public firm CEOs completed an undergraduate degree at a top-twenty university….”

Lots of good little factoids, “…Relative to the rest of the sample, we find top twenty (ten) graduates are 2%(7%) more likely to start a firm with twenty of more employees, 5% (13%) more likely to receive venture capital funding, 2%(3%) more likely to be acquiredand 1%(3%) more likely hold at least one patent…”

You’ll like this quote, “…Rather, our findings provide the best evidence to date that a supply of potential entrepreneurs employed in the workforce today, especially in the top finance, consulting, and technology firms…”

Their takeaway, “…Conventional wisdom suggest workers with the greatest ability to select into entrepreneurship prior to any labor shocks. Any firms created due to local employment shocks will therefore instead be a stop-gap measure until the founder can rejoin the workforce. Forced entrepreneurs would then disproportionally lead to the creation of low-quality firms.Yet, in contrast to these arguments, we find that businesses formed by entrepreneurs entering weak labor markets perform equally, and often better, than those who enter strong labor markets.Upon further examination of the data,we find (i) entrepreneurial success is driven by the top of the income distribution and (ii) these same workers are disproportionately impacted by labor shocks; as a result, labor shocks can increase entrepreneurial quality. Our results therefore provide the best evidence to date that even best entrepreneurs were not destined to select into entrepreneurship, and that there are many wage workers today with high entrepreneurial potential…”

Core findings, “…that a 1 percentage point increase in the national unemployment rate is associated with a (i) 0.5 percentage points decrease in likelihood of employment with a top consulting, finance, or technology firm, (ii) 0.4 percentage point decline in the likelihood of employment with a firm in the Russell 1000 Index, and (iii) 0.3 percentage points decrease in a high-wage industry. 6 In our baseline analysis, we estimate a 1 percentage point increase in the national unemployment rate increases the likelihood of entering entrepreneurship by 0.2 percentage points compared to individuals of the same gender graduating from the same institution and major in the prior year. Given a baseline entrepreneurship rate of 2.2% at graduation, a one standard deviation increase in unemployment increases entrepreneurship by 15% relative to the mean. Since differences in entrepreneurship rates remain steady three years after graduation, transitory labor shocks result in the creation of firms that would otherwise not exist in the economy. Last, we find the majority of these firms are in the financial, professional, and technical services sectors, and this industry composition can be largely explained by the skillsets of the founding entrepreneurs. Startups in our sample are more successful than average startups in the economy. We find that 15% of our startups receive venture capital, 4.5% are eventually acquired, and 6% receive at least one patent.7 To confirm labor shocks can impact the entry of successful and innovative ventures, we show that an increase in the unemployment rate increases the rate of (i) large employer firms, (ii) venture-backed firms, (iii) patent-holding firms and (iii) acquired firms…”

They authors built a new dataset, “…We compile a novel data set to conduct our analysis. Our data includes profiles of 640,000 recent college graduates, representing 3% of all workers graduating from a U.S college between 1996 and 2014. Though the workers from our sample graduate from over 2,200 undergraduate institutions, we overdraw graduates of highly-selective universities since these colleges disproportionately graduate successful entrepreneurs. For instance, Massachusetts Institute of Technology, Stanford University, Northwestern University, Cornell University, University of California at Berkeley, and University of Michigan each graduate 1% or more of the employee sample. We identify over 36,000 founders of firms and our sample includes highly-successful entrepreneurs including the co-founders of Airbnb, Dropbox, Instagram, Khan Academy, Paypal, Square, Yelp, and Youtube, among others. In total, 141 of the startups in our sample completed successfully an IPO….Specifically, we compare a worker graduating from a U.S. undergraduate institution to a separate worker of the same gender graduating from the same institution with the same major, but in the two prior years...labor market conditions at the time of graduation have long-term effects on wages even among the most prestigious academic institutions...”
They focused on the top schools, “….focus on the workers in our sample with the greatest ex-ante likelihood of becoming top-earners: those graduating from the twenty (or ten) most selective undergraduate institutions in the U.S.8We provide three findings. First, analyzing over 115,000 elite workers, we find labor shocks on graduates from selective institutions disproportionately increase the likelihood of underemployment. Second, in response to a labor market shock, we find these workers are more likely to enter entrepreneurship relative to other graduates. Third, we confirm entrepreneurs from the most selective institutions are significantly more likely to start high-quality firms relative to the rest of the sample…”

They make an effort to control for personal characteristic by comparing workers by undergraduate institution, sex and major versus their comps in a non-weak job market.

Counterintuitive Research on the elasticity of entrepreneurs suggests talent is the limiting constraint. Research finds firms founded by individuals who were "forced" into entrepreneurship by a poor job market (ie they were unable to get a prestigious or high earning job at time of graduation) are more likely to survive, innovate, and receive venture-backing.

Ed Comment: "Not surprising. Smart guys do better and shocks increase the likelihood of entrepreneurship, presumably at all skill levels. Does it say how much a shock affects the top relative to those below the top? Your write up only said "more." "Second, in response to a labor market shock, we find these workers are more likely to enter entrepreneurship relative to other graduates.""
Ben Comment:Here are the magnitudes: "..."we demonstrate that a 1 percentage point increase in the national unemployment rate is associated with a (i) 0.5 percentage points decrease in likelihood of employment with a top consulting, finance, or technology firm, (ii) 0.4 percentage point decline in the likelihood of employment with a firm in the Russell 1000 Index, and (iii) 0.3 percentage points decrease in a high-wage industry."..."

Addressing how much does this change the rate of entrepreneurship: "we estimate a 1 percentage point increase in the national unemployment rate increases the likelihood of entering entrepreneurship by 0.2 percentage points compared to individuals of the same gender graduating from the same institution and major in the prior year. Given a baseline entrepreneurship rate of 2.2% at graduation, a one standard deviation increase in unemployment increases entrepreneurship by 15% relative to the mean. Since differences in entrepreneurship rates remain steady three years after graduation, transitory labor shocks result in the creation of firms that would otherwise not exist in the economy.”

Addressing how successful these starts ups are (good companies): "We find that 15% of our startups receive venture capital, 4.5% are eventually acquired, and 6% receive at least one patent. To confirm labor shocks can impact the entry of successful and innovative ventures, we show that an increase in the unemployment rate increases the rate of (i) large employer firms, (ii) venture-backed firms, (iii) patent-holding firms and (iii) acquired firms.”

And "Compared to firms started during periods of low unemployment, we find forced entrepreneurs have the same employment size, and identical probability of successfully completing an IPO. Strikingly, firms started by entrepreneurs who entered the labor market during distressed labor markets are actually more likely to survive, obtain venture capital financing, produce a patent, or be acquired.”

Addressing the composition of total entrepreneurs: "To be clear our results do not imply recessions increase the level and quality of en- trepreneurship. In fact, the overall effect of recessions on the rate and composition of entrepreneurship is negative (Moreira, 2016)” - more companies are started during recessions but most of these are started by people at the low end of the talent pool. The overall quality of these businesses is low. BUT - the ones started by the high end guys (who were forced to start their own because google wasn’t hiring) is very high.

Regarding how many people get top jobs - it looks like it’s about 9% in their data (6th row).

Forced Entrepreneurs: Comments Image 1


More detail on company quality: "First, 23% startups remain in business as of 2019, while 9.4% (5.1%) currently employ at least twenty (fifty) workers. Second, at least 6% of the firms created at least one patent, including 2.5% that patent within five years of establishment. Third, 15% received VC funding throughout their firm life, and 12% receive VC funding within five years of establishment. Fourth, 4.5% were acquired by a separate firm. Lastly, 141 startups completed successfully an Initial Public Offering (IPO), representing 5.7% of all IPOs between 1999 and 2012”

Comparing the top 20 to everyone else: "Even after controlling for gender and major, we find (i) graduates of the top twenty institutions are better entrepreneurs than other grad- uates of the sample, and (ii) graduates of the top ten institutions are substantially better entrepreneurs than graduates of the next ten institutions. Relative to the rest of the sample, graduates from top twenty (ten) institutions are 2% (7%) more likely to start a firm with twenty of more employees, 5% (13%) more likely to receive venture capital funding, 2% (3%) more likely to be acquired and 1% (3%) more likely hold at least one patent.”

How badly are top folks shocked? Turns out, a lot: "We estimate that a 1 percentage point increase in the unemployment rate decreases the rate of employment with top finance, consulting, or technology firm by 1.3 percentage points; in comparison, we estimate a 0.24 percentage points decline for all other workers, and this difference is statistically significant at the 1% level. Based on this measure, we find evidence that graduates of highly-ranked undergraduate institutions are particularly exposed to labor market shocks.” However - this defines shocks interestingly. It seems like these people will probably end up ok, just maybe not at top top firms.

So they go on: "We estimate a 0.76 percentage point decrease in the likelihood of joining a Russell 100 employer (compared to a 0.35 percentage point decline for all other workers) and a 0.93 percentage point decrease in employment within a high-wage industry (compared to a 0.072 percentage point decrease for other workers). In both cases the differences in coefficients are statistically different, providing additional evidence that employment shocks are more costly for graduates of high-ranked undergraduate institutions."