“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
Ed’s CBS Interview Notes
Does capitalism still work for the middle- and working-class?
With 3 to 4 times as many low-skilled workers per high-skilled worker as Northern Europe, America would be significantly richer if we enjoyed the same distribution of talent. And Europe would be poorer if it wasn’t benefiting from America’s disproportionately large contribution of innovation, defense spending, and healthcare profits. Apple alone is worth more than the 30 largest companies in Germany. And America is producing 5 times as many billion-dollar startups. Europe has contributed shockingly little innovation.
We shouldn’t take these enormous differences for granted. Trying to increase prosperity by increasing the size of government is more costly and risky than its advocates admit, although its detrimental effects will be felt gradually over time.
Why has America alone been so much more successful than other high-wage economies?
For a variety of reasons—exposure to cutting-edge ideas at companies like Google and Facebook, for example—the rewards for successful risk-taking are greater in America than Europe. With larger rewards, talented Americans have gotten better training, workedlongerhours, taken more entrepreneurial risk, and produced a lot more innovation than their counterparts in Europe and Japan. Their success has gradually built companies, like Google, Microsoft, Intel, and Amazon, that can successfully mine the technological frontier. Greater exposure to the technological frontier arms talented Americans with better ideas. In turn, good ideas increase the payoff and likelihood of success, which increases the risk-taking that produces innovation and growth in today’s information-oriented economy.
Look at Apple, for example. Apple alone is worth more than the 30 largest German companies combined. No other high-wage economy has created companies like Apple, Microsoft, Google, Facebook, Amazon, and Intel. America is producing 5 times as many billion-dollar startups as Europe. US productivity, as measured by GDP per hour worked, has grown 50% faster than Northern Europe since 2000, and three-times faster than Southern Europe with demographics similar to America. Europe’s productivity growth would be slower still without the outsized contribution of American-made innovation. These are enormous differences.
Even liberal economists agree that competition forces business owners to produce $5of value for others—customers, employees, and suppliers—to earn a dollar for themselves. We can discourage growth by more heavily taxing the 50 cents business owners keep after taxes, or maximize incentives for risk-taking by restraining the never-ending increase in the share of the economy consumed by government.
If higher taxes slow growth, why wasn’t growth slower in the 1990s, in California, and in the 1950s?
The internet was commercialized in the 1990s. It was like creating the telephone. It increased the payoffs for successful risk-taking more than the Clinton Administration’s tax increase reduced them. Higher payoffs increased entrepreneurial risk-taking. Successful innovation accelerated growth and raised middle-class incomes.
Similarly, Silicon Valley increases the payoff for entrepreneurial risk-taking more than higher California state taxes reduces them. Higher payoffs increase entrepreneurial risk-taking. California’s high taxes slow growth at the expense of the rest of the world.
Since the 1950s, we have transitioned from a more command-and-control manufacturing economy with fast productivity growth to a hard-to-manage decentralized service economy with slower productivity growth. We also enjoyed a large one-time increase in productivity growth from saturating the population with education.
Arguments for higher marginal tax rates, such as Diamond and Saez’s caveat-filled justification for a 70 percent marginal rate, admit that estimates of the effects of higher taxes on growth “should reflect not only short-run labor supply responses but also long-run responses through education and career choices.” They concede, “we unfortunately have little compelling empirical evidence to assess whether taxes affect earnings through those long-run channels.” Nevertheless, they assume the long-run effects are nonexistent despite growing differences between America and the rest of the high-wage world.
Nor can economists specify optimal tax policy independent of the expected payoffs for risk-taking, which Diamond and Saez ignore. High payoffs may motivate risk-taking despite high tax rates, whereas low payoffs may discourage risk-taking no matter the tax rate. At the same time, tax rates high enough to discourage risk-taking may have costly consequences when expected payoffs are high, like they appear to be in America, and have little, if any, consequences when expected payoffs are low, as they appear to be in Europe. Diamond and Saez ignore these tradeoffs.
Decades of risk-taking have gradually built American institutions, like Google, that can successfully mine the technological frontier, train workers, spinoff valuable ideas, and raise the expected payoffs for risk-taking. Raising taxes reduces the payoffs for successful risk-taking. Lower payoffs reduce incentives for risk-taking, the demand for effective training, and the supply of valuable trainees, which gradually slows growth. For the same reason, Europe won’t immediately grow faster if it cuts taxes because it hasn’t taken the risks needed to produce institutions that can successfully mine today’s technological frontier. It takes decades of successful risk-taking to gradually create these assets. Optimal tax policy must maximize utility in the long run, not the short run. Analysis that considers the long-term effect of tax rates on innovation and growth derives revenue-maximizing marginal tax rates of 30 percent or less.
Won’t talented workers keep producing regardless of the tax rate?
Professors have the intelligence to forsake money and to try to earn status through their intellectual achievements. They mistakenly assume others are like them. But the most productive workers aren’t smart enough to distinguish themselves through their intellectual achievements. They achieve status by earning money, power, and fame. Unlike Einstein, who pursued science to “escape from everyday life with its painful crudity and hopeless dreariness,” the most productive workers embrace the unrelenting demands of serving customers more effectively than competitors. Like athletes, they endeavor to outwork their competitors where small differences in performance produce large differences in results.
Were status merely relative, Europe’s economy would be as competitive as America’s. America and Russia won’t compete vigorously to produce the military difference between a peashooter and a flyswatter, for example, even though the relative difference may be large. Unfortunately, most differences aren’t large enough to matter, which is why the most talented students tend to work harder than average students. Unlike average students, the same amount of work by talented students produces large differences in results. Like evolution, allowing individuals to maximize the absolute size of their differences maximizes incentives and the ferocity of competition. Vigorous competition has produced the differentiated success of the U.S. economy.
Aren’t rich people just lucky?
Luck is the flip side of risk. Success bubbles up randomly from large pools of failure. If people knew which innovations worked best, those innovations would already be produced. No one has a clue. We don’t need payoffs to motive success; we need them to motivate large pools of failure.
Like the byproducts of evolution, things tend to exist for sound reasons. It’s hard to find improvements and nearly impossible to get rich. Believed improvements are overwhelmingly the byproducts of mistaken analysis of a world that is far too complex to analyze.
This is the core problem with liberalism. It’s proponents naively assume they know better, grossly underestimate the risk of change, and, as such, over-confidently propose singular government changes that have not evolved into existence through survival of the fittest by adapting to fiercely competitive alternatives—a process that is essential for optimization and robustness.
But why does Bill Gates need to earn so much?
It’s true billionaires, like Steve Jobs and Bill Gates, earned a lot more money than they expected to earn when they initially took risk. But their outsized success motivated an army of talented risk-takers to follow in their wake. An army of risk-takers allows a small amount of innovation to bubble up unexpectedly from large pools of failure. Their success has produced higher middle-class incomes and faster employment and income growth in America than in all other high-wage economies.
Basic economic theory predicts risk-taking will expand until expected after-tax returns drop below a minimum threshold. Indicative of the theory, averaging gains and loss, venture capital and other private equity investments have scarcely outperformed public markets. Risk underwriters must compensate properly trained talent for bearing enormous company-specific risks, even though financial investors can diversify away these risks. Moreover, every investor knows that returns are disproportionately earned by a handful of homerun investments. Capping returns significantly diminishes expected returns. Lower expected returns diminish risk-taking and the demand for training, which gradually slows growth.
Why would we want to discourage innovators, like Jeff Bezos, from creating large companies, like Amazon, by limiting how much money they can earn? Large companies are more competitive. They invest more, spend more on research & development, grow productivity faster, pay higher wages, are more resilient to downturns, and create economies of scale critical for mining the technological frontier that spins off valuable ideas that grow the economy faster—all characteristics we want America’s economy to achieve.
Unlike Bill Gates, haven’t most of the rich earned their money through cronyism?
Proponents of income redistribution insist rich people are earning their money unfairly. Were it true, we could redistribute income without slowing growth. But if America was misallocating resources, America’s growth—the country with the greatest inequality—should be slowing relative to countries like Europe with more equally distributed incomes. America’s growth is faster than Europe’s. It’s true large companies are gaining market share, but they are more productive, invest more, and pay higher wages—the opposite of misallocated resources.
Aside from the slow-growing automotive sector, America’s 200 largest companies are investing twice as much as their European counterparts in R&D. The US economy is investing nearly 25% more in intangible assets, such as software and training, than Europe. America is investing about eight times more venture capital per dollar of GDP. Rising cronyism isn’t evident in any of these revealing comparisons.
Were cronyism rising, we should also expect to see an increasingly entrenched status quo. We see the opposite. The turnover of the Fortune 500, CEO tenures, and the Forbes 400 richest Americans have also increased. Most of America’s growth is coming from technology, which has been in turmoil. Excluding Microsoft, the 15 largest NASDAQ technology companies at the peak of the internet bubble have lost almost 60 percent of their market value. Meanwhile, half of today’s largest 15 NASDAQ companies, with a combined market capitalization of $3.5 trillion, were worth less than $100 billion in 2000. Formerly dominant phone makers—Palm Pilot, Motorola Nokia, Ericsson, BlackBerry—are shadows of themselves. There are now over 100 manufacturers making thousands of models. Of the 500 companies in the Fortune 500 in 1955 only 51 remain.
Oligarchical control of the government is obviously suboptimal. With politicians trying to buy the majority at the expense of the minority; an uninformed electorate easily mislead by propaganda; consumers eager to consume income that would otherwise be invested; government projects unrivaled by competition; and special interests trying to influence lawmakers, Churchill recognized democracy was the worst form of government except for all the others. It’s hardly obvious that more influence by a minority of investors necessarily endangers democracy and threatens prosperity at the margin, especially when the majority of voters would eagerly vote to tax, redistribute, and consume income that would otherwise underwrite risk-taking, which is essential for growth.
It is true that lobbying is growing. But lobbying is often like steroids in sports where competitors largely fight each other to a draw.
America’s founders didn’t write the Constitution to protect us from capitalism. They wrote it to protect us from government.
What about Wall Street?
The rest of the world floods America with risk-averse savings. That’s why interest rates are so low. An inflow of savings creates trade deficits. When we buy more from offshore workers than they buy from us, Americans lose jobs. These workers—chiefly low-skilled workers—aren’t rehired until someone borrows and spends the savings. To reach full employment at the highest possible wages, Wall Street must find uses for these savings unless the government runs fiscal deficits. Wall Street’s job is much more important to the prosperity of America’s workers than Wall Street’s critics admit.
With little use for this flood of risk-averse offshore savings, Wall Street engineered the ability of homeowners to borrow against the embedded equity in their homes and use the proceeds for other purposes. Unfortunately, homeowners unwisely consumed rather than invested their embedded equity. Lawmakers and homeowners applauded this innovation, until they realized that a flood of risk-averse savings puts America’s growth under enormous pressure to borrow and spend and destabilizes an inherently unstable banking system—a lesson we learned decades earlier but had long since forgotten by the time of the financial crisis.
Why is there more inequality in America than Europe?
For a variety of reasons—exposure to cutting-edge ideas at companies like Google and Facebook, for example—the rewards for successful risk-taking are greater in America than Europe. With larger rewards, talented Americans have gotten better training, worked long hours, taken more entrepreneurial risk, and produced a lot more innovation than their counterparts in Europe and Japan. Successful risk-taking has produced a lot more innovation and faster growth in America. Europe and Japan would be much poorer without the disproportionate contributions of America than America would be without them.
Growth increases wages when the supply of workers is constrained. Otherwise, it increases employment. Automation, trade with low-wage offshore economies, and low-skilled immigration have given America access to an unconstrained supply of low-skilled workers and substitutes. America has responded by growing employment twice as fast as Europe since the 1980s. If we want to increase middle- and working-class incomes, we need to constrain the supply of workers, although this will come at the cost of slower growth.
In today’s knowledge-driven economy, with an abundance of low-skilled labor and offshore capital, properly trained talent constrains growth. We should train more high-skilled workers if we can, or recruit them from abroad if we can’t.
Why has inequality grown?
Most economists find popularized measures of wealth and income inequality to be highly misleading. The true cost of a person to society is how much they consume, not what they own or earn—much of which is taxed, invested, or given away, which helps others.
Accounting for government benefits, taxes, and investment, the rich (the 90th percentile), albeit not the richest 1%, consume about 4 times more than the poor (the 10th percentile). This has barely changed since the 1960s, even though the rich have earned more, saved and invested more, and subsequently own more.
As the economy grows larger relative the people who compose it, like companies, individuals who achieve economy-wide success will grow larger relative to the median worker. Hopefully, maximizing middle- and working-class incomes is our objective; not redistributing income for its own sake.
Isn’t the middle class being hollowed out?
Pew Research Center claims America’s middle class has declined from 61% in 1970s to 50% percent today. Seven of the 11 point declined came from families who move upward in income. Four of the 11 points from families that moved down. But Hispanic immigrants account for 3 of the 4 points of decline. So native-born Americans have enjoyed a seven point upward shift in income offset by a single point of decline. Other studies reach similar conclusions.
Why aren’t essential workers paid more?
Water is essential, and yet the price of water is low because of its abundance. The same is true of lesser-skilled workers. Low-skilled immigration, $3 an hour Mexican and Chinese wages, and automation have created an abundance of low-skilled labor. This abundance holds down low-skilled wages.
If we raise the price of water— people will use less of it. The same is true of workers. Employers don’t hire workers; customers hire workers. Employers hire workers on behalf of customers. If we raise the minimum wage to $25 an hour, customers won’t hire low-skilled workers who can’t produce $25 of value for them. And the incomes of those workers will fall. Countries that have tried to legislate prosperity have lower middle-class incomes.
People are hurting. Don’t we have a moral obligation to help them?
People overwhelmingly agree that we should help those in need; and we have. But we shouldn’t do it blindly. People demand more spending without knowing how much we spend—per person; as a share of the economy; and relative to Europe. Last year—not counting the $1.6 trillion the government spent on retirees—America spent $1.25 trillion helping the poor. That’s enough to give every person in the bottom 20% under the age of 65, $22,500, or $90,000 per family of four—50% more than America’s middle-class family earns, who we’ve already seen are the richest middle class in the world by far.
Unfortunately, we don’t give a large share of that money to the poor. Instead, lawmakers use it to buy votes. Nevertheless, America spends more after tax helping the poor than the richest countries in Europe, who, unbeknown to most, tax their poor with 20% sales taxes.
So, while most everyone agrees that we should help the poor, not everyone agrees that we need to increase government spending from 36% of GDP (including state and local government)—which is expected to rise to 41% of GDP as baby boomer retire—to do it.
Countries that have increased government spending as a share of GDP have significantly smaller and slower-growing middle-class incomes. Unfortunately, we can’t legislate prosperity. We have to earn prosperity the old fashion way—with hard work, investment, good ideas, and entrepreneurial risk-taking.
Unfortunately, we can neither legislate nor tax our way to prosperity. Countries that have tried have lower middle-class incomes and slower growth rates. There is clearly more value in increasing the $5that business owners must produce to put a dollar in their pocket, than in taking more of the 50 cents they keep after taxes. To minimize taxes and maximize incentives to complete the training and take the risks necessary to grow the economy, we must reallocate spending to accomplish objectives, such as helping the poor more effectively and investing in infrastructure and R&D, rather than increasing government spending as a share of GDP.
In today’s knowledge-driven economy, with America’s abundance of low-skilled labor and offshore risk-averse savings, properly trained talent constrains growth, and with more valuable opportunities for high-skilled worker than creating products and jobs for low-skilled workers. We should train more high-skilled workers if we can, or recruit them from abroad if we can’t.
Growth increases wages when the supply of workers is constrained. Otherwise, it increases employment. If we want to increase middle- and working-class incomes, we need to constrain the supply of lesser-skilled workers relative to higher-skilled workers, although this will slow growth relative to what it could have been.