Squeezing the Worker

No society can be flourishing and happy, of which the far greater part of the members are poor and miserable.

— Adam Smith, The Wealth of Nations

(Excerpt from The Myth of Capitalism - to be published November 2018).

This book started out as a simple detective story: Who killed your paycheck?

When we’re not writing books, we spend our hours looking at economic charts of leading economic indicators. Sometimes, we even think about them when we shave or in the bath, where we hope for a eureka moment. Our clients pay us to tell them where the economy is heading and how they should invest. They want to know answers to important questions like: Will the United States go into a recession? Will China experience a big debt crisis? Will Italy leave the euro? Will US wages rise?

Our clients are not economists. They are entrusted with the savings of pensioners and the insurance premiums families have paid. They’re investors managing people’s savings, and it matters to them whether inflation is rising, whether unemployment is falling, and whether profits are rising. These factors affect the investments of pension funds, insurance companies, and mutual funds. They want to avoid major market crashes, and benefit when markets are good.

Most of the charts we use to advise our clients go back decades and are based on sound fundamental relationships, so they never need changing. If building permits are strong, the housing sector will flourish, and the economy will be strong. If the People’s Bank of China is hiking interest rates, the Chinese economy will likely slow down. If the unemployment rate is falling, it is likely that wages will go up, as companies must competitively bid for workers.

Once in a while, the tools and charts we use to infer future trends will appear “broken” and will stop working. Either something has changed in the world, or we need to figure out why we’re wrong.

One chart in particular was bugging us. At meetings with money managers in mid-town Manhattan, we would present our reports while they flicked through our charts. We would stare out at Central Park while they scoured the lines that moved up and down. They would scroll through the pages, but they always stopped at the page with our chart of US wages (Figure 1). It had barely gone up in the past nine years.

Figure 1 - Average Hourly Earnings versus VP US Wages Leading (Advanced 15 Months)

Source: Variant Perception.

Our US Wages Leading Indicator is supposed to lead wages very closely and tell us whether US workers will get a raise or not. It gathers data on how tight the labor market is, how low initial unemployment claims are, and other factors affecting a worker’s ability to get a raise. Across half a dozen business cycles, it has told us when workers would be paid more and when corporate profits would be crimped.

Back in Manhattan, our client questioned us: “Why is your indicator telling me that wages are going up, when I don’t see that in the data? It looks broken,” our client said. “Why are you wrong?”

“Trust me, wages will turn up. Just wait a big longer. The lead time is very long.”

The first time this happened, we genuinely thought wages would rise and our indicator would be right. After all, if the job market is tight businessmen say they will increase wages. Yet workers’ pay barely went up.

This leading indicator had worked flawlessly for decades, but over the past few years it started to behave in a strange way. Our index kept moving higher, yet workers’ pay never budged. Meanwhile, corporate profits were rising to all-time highs and stayed there. In fact, corporations have never had it so good.

If capitalism were a game, it would be: Workers 0 – Corporations 1.

As months passed and the wage data came out, it became clear. Our indicator was not working. Corporate profits seemed to be defying the gravity of competition. We had missed something very big happening in the economy, and we had no idea what was causing wages to stay so low.

Something was very broken. The rules of the game had changed for American workers. This book became our attempt to answer why.

Today everyone in the world thinks of Silicon Valley as the cradle of technology, but in the 1950s, people would have thought you were crazy if you set up a tech company in California. Santa Clara County was little more than apple trees.[i] Most renowned technology companies had headquarters in Massachusetts along Route 128, near the research hubs of MIT and Harvard.

William Shockley was as close to a rock star as it got in the science world. He was a Nobel Prize winner who had co-invented the transistor. When he moved to Palo Alto to found Shockley Semi-Conductor Laboratory, people thought he had gone mad. It was far removed from Route 128, but he had his reasons. He had grown up in the area, and he wanted to return home to help his ailing mother.

Shockley hired an all-star cast to join him. They were all experts in physics, metallurgy and mathematics, and they abandoned the East Coast to work with Shockley to commercialize the transistor. Robert Noyce, one of the hires, said getting the call from Shockley was like picking up the phone and talking to God.

Shortly after arriving, however, they discovered that Shockley was an erratic and difficult boss. He was a genius, but a jerk, and not just your ordinary, run-of-the-mill jerk. He was an epic egomaniac. When his colleagues at Bell Labs discovered the transistor, he tried to claim sole credit. Later in life, he spent his time espousing a racist eugenics agenda, promoting a high-IQ sperm bank, and losing all contact with his children. He was, by most accounts, a horrible boss.

Within a year of joining Shockley, the new hires sat around a breakfast table at the Clift Hotel to plot their escape. Although they were America’s most brilliant scientists and engineers, they were distinctly unhappy working for Shockley. In a brazen display of disloyalty, they decided to leave and found a new company: Fairchild Semiconductor. Later dubbed the “Traitorous Eight,” they signed one dollar bills in place of formal contracts – a symbol of nonconformity.

Many consider this act of employee treachery as the definitive moment of Silicon Valley’s creation, though the term would take another 10 years to enter the public lexicon. The defection set a precedent of ‘can-do’ entrepreneurialism and loyalty to lofty ideas, rather than individual firms and egos.[ii]

The ringleader was Noyce, only 29 at the time, and the group’s resident transistors expert. Even he had abandoned “God.” Eventually Noyce and his colleague Gordon E. Moore outgrew Fairchild, and again poached employees to start Intel. In 1971, only three years after founding Intel, Noyce made history yet again with the invention of the Intel 4004, the world’s first microprocessor. He had invented the beating heart of the modern computer.

Silicon Valley owes its success to many things – access to capital, proximity to Stanford (one of the best universities in the world), and being close to a vibrant city like San Francisco. But what helped make it the innovation capital of the world is rarely discussed: California is one of the few states where noncompete clauses in work contracts are completely nonenforceable. In other words, employees have full rights to leave and work for a competitor.

In many other states, when employees join a firm, they may be asked to sign a noncompete agreement as a condition of their employment. The terms vary considerably, but the basic idea is that if you are fired or quit your job, you cannot work for a competitor within the same industry for a certain period of time – which can range from a few months to years. These clauses deprive workers of their livelihoods if they make it difficult for them to leave jobs and bargain for higher wages elsewhere.

In 1872 California made it illegal for employees to be bound to a specific employer, allowing them to move freely between jobs and firms. This state law is still in effect nearly 150 years later. The lack of noncompetes is a primary reason why the Valley achieved such tremendous success. To this day, Boston continues to lag Silicon Valley in the commercialization of new technologies.

Imagine where the Valley would be today if Noyce were prevented from defecting and founding a new firm with his colleagues. What if Wozniak had never left Hewlett Packard to join Steve Jobs? Think of the history of technology. Imagine where we would be today if Nikola Tesla were prevented from leaving Thomas Edison.

Silicon Valley’s history demonstrates that respect for worker talent was prized above strict company loyalty. This lead to a malleable ecosystem, where good ideas spread quickly from company to company and innovators were free to choose their own fates. Professor AnnaLee Saxenian, author of many books on the tech industry, points out that, “In the early days engineers would say, ‘I work for Silicon Valley.’ And the idea was that they were advancing technology for a region, not any single company’s technology. We often think in the U.S. that people or companies create success, but what Silicon Valley shows us is that often it’s communities of people across a region.”[iii]

If Noyce thought Shockley was God in the early 1950s, Steve Jobs idolized Noyce in the 1970s. When Apple was starting, Noyce was already a legend with Intel. “Bob Noyce took me under his wing,” Jobs said. “He tried to give me the lay of the land, give me a perspective that I could only partially understand.” Jobs continued, “You can’t really understand what is going on now unless you understand what came before.”[iv]

Although Jobs worshipped Noyce, he failed to give his own Apple employees the same freedoms that allowed Noyce’s best innovations to flourish. In 2014 it came to light that Jobs had been preventing employees from moving to other companies. Silicon Valley was founded on freedom of mobility for workers, but the tech giants – Apple, Facebook, Google, Adobe and many others – were caught in “gentlemen’s agreements” to not poach each other’s employees. Staff brought the case forward claiming that these pacts made it difficult to market their skills and that they also suppressed their salaries.

As part of the lawsuit, emails came out in court between Steve Jobs and Eric Schmidt, the CEO of Google, “I am told that Google’s new cell phone software group is relentlessly recruiting in our iPod group. If this is indeed true, can you put a stop to it? Thanks, Steve.” In another email, Larry Page from Google sent a distressed message saying Steve Jobs had threatened war if a single one of his staff were hired.[v]

In the end, the agreement not to poach went out across Silicon Valley. Google, Adobe, and others developed Do Not Hire lists. This was clear collusion, and the tech firms were forced to pay a $324.5 million fine for their illegal noncompete pact.[vi]

Some readers may find it hard to feel sorry for highly paid software engineers, but the problems of noncompetes do not end there. What is particularly insidious is that these contractual arrangements are creeping into the entire economy and hurting the poorest the most.

These restraints on trade have spread like an epidemic. Today noncompetes cover almost 18% of the entire American workforce.[vii] Nearly 40% of workers have signed one in previous jobs.[viii] Only California and three other states (Montana, North Dakota and Oklahoma) totally ban noncompete agreements in the United States.

Lawyers sometimes argue that noncompete clauses help protect trade secrets, which help companies to innovate. While it might be understandable for firms that earn most of their revenue from intellectual property to ask key employees to sign noncompetes, is there any good reason to ask camp counselors, janitors, and personal care workers to sign these agreements? There are already federal laws dedicated to protecting IP[ix] and, today, even those who clearly do not possess trade secrets are made to sign them, including 15% of workers without a four-year college degree and 14% of people earning less than $40,000 (Figure 2).[x]

Figure 2 - Percentage of Workers with Noncompete Agreements, by Group

Note: These estimates are preliminary and may differ from subsequent versions of the cited page.

Sources: US Treasury; Data: Starr, Bishara, and Prescott.

These employment clauses are found in a staggering percentage of America’s largest fast-food chains with minimum wage employees. Chains like Burger King, Carl’s Jr., Pizza Hut, and, until recently pressured in 2017 to drop them, McDonald’s. These no-hire rules affect more than 70,000 restaurants – more than a quarter of the fast-food outlets in the United States – according to Alan B. Krueger, who is an economist at Princeton University.

The fast-food industry has been one of the biggest sources of job growth since the recession. More than 4.3 million people are now dipping fryer baskets, which is a 28 percent increase since 2010. The growth in fast food employment is almost double the increase in the overall labor market, according to the most recent data from the Bureau of Labor Statistics.

Arguments about protecting intellectual property are outlandish, given there are relatively few trade secrets in flipping burgers or taking orders. Furthermore, what does Pizza Hut have to lose if one of their store workers decides to work for another Pizza Hut across town? There are no corporate secrets to be lost. The answer is simple: the fewer options workers have, the less freedom they have to find a company that might pay a higher wage. The only function of these rules is to limit worker mobility and diminish their ability to bargain for wage increases. This is modern-day feudalism, and workers have become vassals to corporate lords.

The truth is that noncompetes help firms that want tight control over employees. They offer little for a given industry at large or for the economy. Worst of all, they harm workers and are disastrous for workers’ wages.[xi] Noncompetes are not unique to the fast food industry, but are also frequent in maintenance, health, and food services. You can see in Figure 4.3 that wages are much lower in states that enforce noncompetes and wages are much higher in states that do not enforce them. It is clear to see that noncompetes depress hourly wages (Figure 3).

Figure 3 - Wages under Noncompete Agreements

Source: US Treasury; Data: Starr, Bishara, and Prescott.

Why do workers sign such terrible contracts in a supposedly open and free market? Often employees do not realize they are signing away their right to work somewhere else, as firms are not legally obligated to disclose noncompete clauses in almost all states. According to a study by economists Matt Marx at MIT and Lee Fleming at Harvard University, barely 3 in 10 workers were told about the noncompetes in their job offer, and in 70% of cases, they were asked to sign them after they had already accepted the offer and turned down any alternatives. Half of the time, noncompete agreements were presented to employees on or after their first day of work.[xii] It goes without saying that this hardly represents a true choice for workers.

Barring workers from moving in search of better opportunities works only in an environment where firms have all the power. Because of industrial concentration, a number of firms now have monopsony power – that is, they are the only buyers of labor. A monopoly means there is one seller, and a monopsony means there is only one buyer.

In a monopsony, workers have little choice in where they work and have little negotiating power for wages with employers. In a healthy economy, many firms would be competing equally for workers and would be incentivized to entice new hires with higher wages, better benefit packages, and few restrictions on their next career moves. But monopsonies make it easier for firms to depress worker wages. The classic example of this is a coal-mining town, where the coal plant is the only employer and only purchaser of labor. Today, in many smaller towns, Walmart is the new coal plant – and is the only retail company hiring.

Today, the story of America is largely the story of two economies – rural and urban. It was not always this way. The antitrust movement of the 1940s not only targeted giant corporations, but was also an attempt to weaken regional centers that had amassed too much power. This largely worked and, by the mid 1970s, there was a fairly uniform American standard of living – being middle class in the Mideast was pretty much the same as middle class in New England. America had achieved an incredible feat – a stable majority middle class that was largely consistent across the nation.

In the 1980s, however, many of the policies that had helped ensure this regional equality were neglected or reversed. A great divide formed between rural and metropolitan areas in the United States. Rural towns were left behind, as gleaming centers of industry increasingly attracted talent by offering high-paying jobs.

The divergence are now alarming. In 1980, if you lived in Washington DC, your per-capita income was 29 percent higher than the average American; in 2013 you would be 68 percent above. In New York City, the income was 80 percent above the national average in 1980 and skyrocketed to 172 percent above by 2013.[xiii] Power and money began concentrating in urban centers across the country as a rural ‘brain drain’ occurred.

Major cities attract diverse talent and many corporations, which must bid competitively for workers. Workers living in these cities make significantly more money than workers elsewhere (Figure 4). There is power in numbers, and nurses who have five metropolitan hospitals to choose from will make more money than those who work in a town with only one hospital.

Figure 4 Rural Areas Are Lagging

Source: Bloomberg, Shift: The Commission on Work, Workers, and Technology.

A recent study by Marshall Steinbaum, Ioana Marinescu, and Jose Azar shows that labor monopsony is not only pervasive across the US, but is especially so in non-metropolitan areas (Figure 5). Again, this makes intuitive sense – smaller towns have fewer employment options. Figure 5 represents commuting zones that have a few dominant companies in each industry and very concentrated labor markets.[xiv] Only the major cities are exempt from this high degree of concentrated employers.

Figure 5. Monopsonies in Labor Markets: Commuting Zones with High Labor Concentration

Source: Marshall Steinbaum, Ioana Marinescu, and Jose Azar.

The labor market outside of big cities has been co-opted by a few large players in each industry. Research by Nathan Wilmer of Harvard University shows that pressure from large corporate buyers decreases wages among their suppliers’ workers. When Walmart or other large buyers demand lower prices, suppliers end up taking out of their employees’ paychecks. Wilmer found that the squeeze on suppliers accounts for around 10% of wage stagnation since the 1970s. Increases in industrial concentration have shifted market power and lowered workers’ wage growth.[xv]

Monopsony for workers explains the curious Trump phenomenon. While almost all political analysts doubted that a real estate developer and reality show star could become president, his success almost seems inevitable when you look at where his votes came from.

The overlap of Trump votes with the highly concentrated counties is very high. Trump knew how to connect with voters, when he spoke about rigged markets. He spoke to the average worker’s fears. In the 2016 election Hillary Clinton won 472 counties that represented 64% of US Gross Domestic Product, compared to the 36% for the 2,584 counties that voted for Donald Trump. In many small towns, a single meat packing company, insurer, hospital system, or big box store owned by a distant company has now replaced locally owned businesses. Trump was tapping into a profound, justified anxiety across the country.

The wage squeeze is even greater if you are in a small town, with a small consumer market, competing against corporations. Monopsony means workers have little choice and little power. In Ohio, Amazon is one of the state’s major employers and 10% of Amazon workers are on food stamps.[xvi] Walmart and McDonald’s are also major culprits with 10,000-plus workers at each firm also relying on food stamps, according to the same study by Policy Matters Ohio.[xvii]

The increasing imbalance of power that firms have as sole purchasers, explains not only why harmful noncompetes are more prevalent, but also why wages are dangerously low, why workers accept forced arbitration against employers, and why they waive rights to class-action lawsuits. Workers on their own are in no position to bargain against monopolists and oligopolists.

Without any form of countervailing power, American workers are left to fight alone. It is a toxic cocktail for the average American. The hangover from increased corporate power is real, with many struggling to meet basic needs. This is not the free capitalism we need or the hope that drove the “traitorous eight” in Silicon Valley decades ago.

Workers deserve better.


[i] “Silicon Valley,” American Experience PBS series, Season 25, Episode 3. http://www.pbs.org/video/american-experience-silicon-valley/.

[ii] Alex Tabarrok, “Non Compete Clauses Reduce Innovation,” June 9, 2014, http://marginalrevolution.com/marginalrevolution/2014/06/non-compete-clauses.html.

[iii] Mike McPhate, “California Today: Silicon Valley’s Secret Sauce,” May 19, 2017, https://www.nytimes.com/2017/05/19/us/california-today-silicon-valley.html.

[iv] https://www.csmonitor.com/Technology/2011/1212/Robert-Noyce-Why-Steve-Jobs-idolized-Noyce.

[v] Jim Edwards, “Emails from Google's Eric Schmidt and Sergey Brin Show a Shady Agreement Not to Hire Apple Workers,” March 23, 2014, http://www.businessinsider.com/emails-eric-schmidt-sergey-brin-hiring-apple-2014-3.

[vi] Barry Levine, “4 Tech Companies Are Paying a $325M Fine for Their Illegal Non-compete Pact,” May 23, 2014, https://venturebeat.com/2014/05/23/4-tech-companies-are-paying-a-325m-fine-for-their-illegal-non-compete-pact/.

[vii] Rachel Abrams, “Why Aren’t Paychecks Growing? A Burger-Joint Clause Offers a Clue,” September 27, 2017, https://www.nytimes.com/2017/09/27/business/pay-growth-fast-food-hiring.html.

[viii] Evan P. Starr, Norman Bishara, and J.J. Prescott, “Non-competes in the U.S. Labor Force,” December 8, 2017, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2625714.

[ix] Ryan Nunn, “Leveling the Playing Field for Workers by Reforming Non-competes,” May 6, 2016, https://www.brookings.edu/opinions/leveling-the-playing-field-for-workers-by-reforming-non-competes/.

[x] Office of Economic Policy, U.S. Department of the Treasury, Non-compete Contracts: Economic Effects and Policy Implications. March 2016. https://www.treasury.gov/resource-center/economic-policy/Documents/UST%20Non-competes%20Report.pdf.

[xi] Ibid.

[xii] Matt Marx and Lee Fleming, “Non-compete Agreements: Barriers to Entry … and Exit?” Innovation Policy and the Economy 12 (April 2012), p. 49, eds. Josh Lerner and Scott Stern. National Bureau of Economic Research. http://www.nber.org/chapters/c12452.pdf.

[xiii] Phil Longman, “Why the Economic Fates of America’s Cities Diverged,” November 28, 2015, https://www.theatlantic.com/business/archive/2015/11/cities-economic-fates-diverge/417372/.

[xiv] Marshall Steinbaum, “How Widespread Is Labor Monopsony? Some New Results Suggest It’s Pervasive,” December 18, 2017, http://rooseveltinstitute.org/how-widespread-labor-monopsony-some-new-results-suggest-its-pervasive/.

[xv] Nathan Wilmers, “Wage Stagnation and Buyer Power: How Buyer-Supplier Relations Affect U.S. Workers’ Wages, 1978 to 2014,” American Sociological Review 83, no. 2 (2018): 213–242, https://doi.org/10.1177/0003122418762441.

[xvi] Jade Scipioni, “10% of Amazon's Workforce in Ohio Is on Food Stamps, Report Says,”

January 8, 2018, https://finance.yahoo.com/news/10-amazon-apos-workforce-ohio-162700977.html.

[xvii] Policy Matters Ohio, “SNAP feeds Ohio,” September 6, 2017,


Recent Posts