A friend reminded me that the Treasury controls more than $500 billion in gold reserves. The President has the legal authority to direct the Treasury to sell these reserves any time he likes. This would not be enough to cover a full year, given the size of the current deficit, but it certainly could delay the X-date by several months.
Anyone have any ideas why selling gold is not in the mix of possible courses of action?
A friend reminded me that the Treasury controls more than $500 billion in gold reserves. The President has the legal authority to direct the Treasury to sell these reserves any time he likes. This would not be enough to cover a full year, given the size of the current deficit, but it certainly could delay the X-date by several months.
Anyone have any ideas why selling gold is not in the mix of possible courses of action?
Read More Leer más Join the discussion Participa en la discusión
Read More Leer más Join the discussion Participa en la discusión
• COVID-19CoronavirusEconomic Crisis and RecoveryCrisis económica y recuperaciónEconomic GrowthEl DesarolloInflationUnited StatesEE. UU.
Read More Leer más Join the discussion Participa en la discusión
• Economic GrowthEl DesarolloInequalityLa DesigualdadTechnology
We have long known that people in policy debates have difficulty with arithmetic and basic logic. We got yet another example today in the New York Times.
The NYT profiled Geoffrey Hinton, who recently resigned as head of AI technology at Google. The piece identified him as “the godfather of AI.” The piece reports on Hinton’s concerns about the risks of AI, one of which is its implications for the job market.
“He is also worried that A.I. technologies will in time upend the job market. Today, chatbots like ChatGPT tend to complement human workers, but they could replace paralegals, personal assistants, translators and others who handle rote tasks. ‘It takes away the drudge work,’ he said. ‘It might take away more than that.’”
The implication of this paragraph is that AI will lead to a massive uptick in productivity growth. That would be great news from the standpoint of the economic problems that have been featured prominently in public debates in recent years.
Most immediately, soaring productivity would hugely reduce the risks of inflation. Costs would plummet as fewer workers would be needed in large sectors of the economy, which presumably would mean downward pressure on prices as well. (Prices have generally followed costs. Most of the upward redistribution of the last four decades has been within the wage distribution, not from labor to capital.)
A massive surge in productivity would also mean that we don’t have to worry at all about the Social Security “crisis.” The drop in the ratio of workers to retirees would be hugely offset by the increased productivity of each worker. (The impact of recent and projected future productivity growth already swamps the impact of demographics, but a surge in productivity growth would make the impact of demographics laughably trivial.)
It is also worth noting that any concerns about technology leading to more inequality are wrongheaded. If AI does lead to more inequality, it will be due to how we have chosen to regulate AI, not AI itself.
People gain from technology as a result of how we set rules on intellectual products, like granting patent and copyright monopolies and allowing non-disclosure agreements to be enforceable contracts. If we had a world without these sorts of restrictions, it is almost impossible to imagine a scenario in which AI, or other recent technologies, would lead to inequality. (Imagine all Microsoft software was free. How rich is Bill Gates?)
If AI leads to more inequality, it will be because of the rules we have put in place surrounding AI, not AI itself. It is understandable that the people who gain from this inequality would like to blame the technology, not rules which can be changed, but it is not true. Unfortunately, people involved in policy debates don’t seem able to recognize this point.
We have long known that people in policy debates have difficulty with arithmetic and basic logic. We got yet another example today in the New York Times.
The NYT profiled Geoffrey Hinton, who recently resigned as head of AI technology at Google. The piece identified him as “the godfather of AI.” The piece reports on Hinton’s concerns about the risks of AI, one of which is its implications for the job market.
“He is also worried that A.I. technologies will in time upend the job market. Today, chatbots like ChatGPT tend to complement human workers, but they could replace paralegals, personal assistants, translators and others who handle rote tasks. ‘It takes away the drudge work,’ he said. ‘It might take away more than that.’”
The implication of this paragraph is that AI will lead to a massive uptick in productivity growth. That would be great news from the standpoint of the economic problems that have been featured prominently in public debates in recent years.
Most immediately, soaring productivity would hugely reduce the risks of inflation. Costs would plummet as fewer workers would be needed in large sectors of the economy, which presumably would mean downward pressure on prices as well. (Prices have generally followed costs. Most of the upward redistribution of the last four decades has been within the wage distribution, not from labor to capital.)
A massive surge in productivity would also mean that we don’t have to worry at all about the Social Security “crisis.” The drop in the ratio of workers to retirees would be hugely offset by the increased productivity of each worker. (The impact of recent and projected future productivity growth already swamps the impact of demographics, but a surge in productivity growth would make the impact of demographics laughably trivial.)
It is also worth noting that any concerns about technology leading to more inequality are wrongheaded. If AI does lead to more inequality, it will be due to how we have chosen to regulate AI, not AI itself.
People gain from technology as a result of how we set rules on intellectual products, like granting patent and copyright monopolies and allowing non-disclosure agreements to be enforceable contracts. If we had a world without these sorts of restrictions, it is almost impossible to imagine a scenario in which AI, or other recent technologies, would lead to inequality. (Imagine all Microsoft software was free. How rich is Bill Gates?)
If AI leads to more inequality, it will be because of the rules we have put in place surrounding AI, not AI itself. It is understandable that the people who gain from this inequality would like to blame the technology, not rules which can be changed, but it is not true. Unfortunately, people involved in policy debates don’t seem able to recognize this point.
Read More Leer más Join the discussion Participa en la discusión
I’m impressed. Most reporters might just know what politicians say and do, but New York Times reporters can apparently know their convictions. That’s what the paper told us in a piece about protests in France over President Emmanuel Macron’s decision to raise the age for retirement benefits from 62 to 64.
“Mr. Macron’s decision to raise the legal age of retirement was based on his conviction that the pension system was unsustainable and that changing the program, with its generous benefits, was essential to France’s economic health [emphasis added].”
It’s good that the NYT could tell us Macron’s decision was based on his convictions and not say, a desire not to increase taxes on rich people to help cover the cost of the retirement system or even on working-age people. The employment rate among prime age workers (ages 25 to 54) is actually higher in France than in the U.S., so there is little reason to believe that a modest increase in taxes would lead them to stop working.
It is worth noting workers in France, unlike many workers in the U.S., actually are seeing increases in life expectancy. This means that even with this increase in the retirement age, workers retiring in future decades may still get to enjoy longer retirements than their parents.
I’m impressed. Most reporters might just know what politicians say and do, but New York Times reporters can apparently know their convictions. That’s what the paper told us in a piece about protests in France over President Emmanuel Macron’s decision to raise the age for retirement benefits from 62 to 64.
“Mr. Macron’s decision to raise the legal age of retirement was based on his conviction that the pension system was unsustainable and that changing the program, with its generous benefits, was essential to France’s economic health [emphasis added].”
It’s good that the NYT could tell us Macron’s decision was based on his convictions and not say, a desire not to increase taxes on rich people to help cover the cost of the retirement system or even on working-age people. The employment rate among prime age workers (ages 25 to 54) is actually higher in France than in the U.S., so there is little reason to believe that a modest increase in taxes would lead them to stop working.
It is worth noting workers in France, unlike many workers in the U.S., actually are seeing increases in life expectancy. This means that even with this increase in the retirement age, workers retiring in future decades may still get to enjoy longer retirements than their parents.
Read More Leer más Join the discussion Participa en la discusión
• BudgetBudget and National DebtHealth and Social ProgramsLos Programas Sociales y de SaludHealthcareUnited StatesEE. UU.
Read More Leer más Join the discussion Participa en la discusión
• Economic GrowthEl DesarolloUnited StatesEE. UU.WorkersSector del trabajo
Last week, David Brooks had a column that was quite literally a celebration of American capitalism. He makes a number of points showing the U.S. doing better than other wealthy countries over the last three decades. While his numbers are not exactly wrong, they are somewhat misleading. (I see Paul Krugman beat me to the punch, so I’ll try not to be completely redundant.)
Brooks points to the faster GDP growth in the United States than in other wealthy countries. As Krugman notes, much of this gap is due to the fact that they have older populations; the differences are much smaller if we look at GDP per working-age person.
However, a big part of the story is also that they have opted to take much of the benefit of productivity growth in the form of more leisure time. In France, the length of the average work year was reduced by 9.1 percent between 1990 and 2021. In Germany, the decline was 13.8 percent. In Japan, average hours fell by 20.9 percent. In the United States, the length of the average work year fell by just 2.3 percent. On this score, there is not much to brag about here.
Brooks also cites data showing that the U.S. has seen more rapid productivity growth than other rich countries over the last three decades. This is true (austerity, which is more widely practiced in the EU than in the U.S., is a great way to kill growth), but it is worth mentioning some complicating factors in these comparisons.
For example, more than one-sixth of our GDP goes to health care. As Krugman points out, life expectancy in the United States has been stagnating and, in fact, declining for the poorer half of the population. That suggests that the share of our output that has gone into health care hasn’t done us too much good. (It’s true the decline might have been larger if not for the increased spending, but I’m not sure we want to brag much that our system is great because it slowed the decline in life expectancy.)
There are other things that get counted in GDP, and therefore productivity, that are of questionable value. For example, the money devoted to preventing school shootings (e.g. stronger doors, security alarms, school security guards, and grief counselors after the fact) all add to GDP and productivity growth. In a context where we have school shootings, these might be worthwhile expenditures, but most people would probably agree that we were better off in 1990 when we didn’t have to spend this money and school shootings were a rarity.
The other point about productivity growth is that we are still living in a slowdown world. In the years from 1947 to 1973, annual productivity growth averaged 2.8 percent. Since 1990 it has averaged 1.9 percent. That isn’t horrible, but it is markedly slower than the Golden Age growth. It is also worth noting that much of this growth was due to the Internet boom from 1995 to 2005 when growth averaged 3.0 percent annually. In the years before and after this period, productivity growth averaged just over 1.0 percent.
As Krugman points out, the benefits of the growth we have seen went disproportionately to those at the top. The median wage grew by 18.9 percent from 1990 to 2022, an average of just over 0.5 percent annually. It’s also worth noting that almost all of this growth took place in the tight labor markets of the late 1990s and from 2015 until the pandemic.
Finally, Brooks has the U.S. share of world GDP being unchanged in the years since 1990. This is likely because he was using GDP measured at exchange rate values. This is a rather arbitrary measure that varies hugely as currencies fluctuate. The U.S. GDP by this measure will also be inflated insofar as countries “manipulate” their currency by buying up dollar-based assets with their reserves.
A more standard measure of economic output for purposes of international comparisons is purchasing power parity GDP. This measure applies a common set of prices to all goods and services produced in each country. This measure tells a very different story.
Source: International Monetary Fund.
As can be seen, the U.S. share has declined from 21.5 percent in 1990 to a projected 15.4 percent this year. I have also included the I.M.F. projections which show a further decline to 14.5 percent by 2028. Perhaps more importantly, China’s share has risen dramatically.[1] It crossed the U.S. share in 2014 and is projected to be 19.3 percent this year. It is projected to rise further to 20.1 percent in 2028, not far below the U.S. share in 1990. In short, China is now the top dog in terms of contribution to work GDP by a fairly large margin.
There are still plenty of good things that can be said about the U.S. economy, but there is perhaps less to celebrate than David Brooks would have us believe.
[1] I’ve included Hong Kong and Macao’s data in the calculation of China’s GDP share.
Last week, David Brooks had a column that was quite literally a celebration of American capitalism. He makes a number of points showing the U.S. doing better than other wealthy countries over the last three decades. While his numbers are not exactly wrong, they are somewhat misleading. (I see Paul Krugman beat me to the punch, so I’ll try not to be completely redundant.)
Brooks points to the faster GDP growth in the United States than in other wealthy countries. As Krugman notes, much of this gap is due to the fact that they have older populations; the differences are much smaller if we look at GDP per working-age person.
However, a big part of the story is also that they have opted to take much of the benefit of productivity growth in the form of more leisure time. In France, the length of the average work year was reduced by 9.1 percent between 1990 and 2021. In Germany, the decline was 13.8 percent. In Japan, average hours fell by 20.9 percent. In the United States, the length of the average work year fell by just 2.3 percent. On this score, there is not much to brag about here.
Brooks also cites data showing that the U.S. has seen more rapid productivity growth than other rich countries over the last three decades. This is true (austerity, which is more widely practiced in the EU than in the U.S., is a great way to kill growth), but it is worth mentioning some complicating factors in these comparisons.
For example, more than one-sixth of our GDP goes to health care. As Krugman points out, life expectancy in the United States has been stagnating and, in fact, declining for the poorer half of the population. That suggests that the share of our output that has gone into health care hasn’t done us too much good. (It’s true the decline might have been larger if not for the increased spending, but I’m not sure we want to brag much that our system is great because it slowed the decline in life expectancy.)
There are other things that get counted in GDP, and therefore productivity, that are of questionable value. For example, the money devoted to preventing school shootings (e.g. stronger doors, security alarms, school security guards, and grief counselors after the fact) all add to GDP and productivity growth. In a context where we have school shootings, these might be worthwhile expenditures, but most people would probably agree that we were better off in 1990 when we didn’t have to spend this money and school shootings were a rarity.
The other point about productivity growth is that we are still living in a slowdown world. In the years from 1947 to 1973, annual productivity growth averaged 2.8 percent. Since 1990 it has averaged 1.9 percent. That isn’t horrible, but it is markedly slower than the Golden Age growth. It is also worth noting that much of this growth was due to the Internet boom from 1995 to 2005 when growth averaged 3.0 percent annually. In the years before and after this period, productivity growth averaged just over 1.0 percent.
As Krugman points out, the benefits of the growth we have seen went disproportionately to those at the top. The median wage grew by 18.9 percent from 1990 to 2022, an average of just over 0.5 percent annually. It’s also worth noting that almost all of this growth took place in the tight labor markets of the late 1990s and from 2015 until the pandemic.
Finally, Brooks has the U.S. share of world GDP being unchanged in the years since 1990. This is likely because he was using GDP measured at exchange rate values. This is a rather arbitrary measure that varies hugely as currencies fluctuate. The U.S. GDP by this measure will also be inflated insofar as countries “manipulate” their currency by buying up dollar-based assets with their reserves.
A more standard measure of economic output for purposes of international comparisons is purchasing power parity GDP. This measure applies a common set of prices to all goods and services produced in each country. This measure tells a very different story.
Source: International Monetary Fund.
As can be seen, the U.S. share has declined from 21.5 percent in 1990 to a projected 15.4 percent this year. I have also included the I.M.F. projections which show a further decline to 14.5 percent by 2028. Perhaps more importantly, China’s share has risen dramatically.[1] It crossed the U.S. share in 2014 and is projected to be 19.3 percent this year. It is projected to rise further to 20.1 percent in 2028, not far below the U.S. share in 1990. In short, China is now the top dog in terms of contribution to work GDP by a fairly large margin.
There are still plenty of good things that can be said about the U.S. economy, but there is perhaps less to celebrate than David Brooks would have us believe.
[1] I’ve included Hong Kong and Macao’s data in the calculation of China’s GDP share.
Read More Leer más Join the discussion Participa en la discusión
Read More Leer más Join the discussion Participa en la discusión
• JobsTrabajosUnited StatesEE. UU.WorkersSector del trabajo
Glenn Kessler, the Washington Post’s long-time fact checker, usually tries to be fair, but he really fell down on the job in a piece evaluating Kevin McCarthy’s claim that Biden’s policies cost families $7,400 a year. As the headline tells readers, Kessler concluded that it “depends on your math.” Nope, it depends on being honest.
As Kessler explained, the number came from E.J. Antoni, a research fellow in regional economics with the Heritage Foundation’s Center for Data Analysis. There are two parts to Antoni’s calculation.
The first part looks at average weekly earnings. Antoni said that average real weekly earnings were 5.1 percent lower in the most recent month than when President Biden took office in January of 2021. This loss came to $2,802.77, over the course of a year. Antoni doubles this figure, assuming two workers per household, which gives him a loss of $5,605.53 due to lower pay.
As Kessler sort of points out, this figure is distorted by composition effects. Millions of lower-paid workers lost their jobs in 2020, which raised the average wage of those still working. We can see that story clearly if we look at the longer picture.
As shown below, average real earnings soared in April of 2020, as millions of workers in hotels, restaurants, and other low-paying industries lost their jobs.
Source: Bureau of Labor Statistics and author’s calculations.
Pay fell as workers in lower-paying industries got their jobs back. This decline did not reflect a fall in the real wages of individual workers but rather a change in the mix of workers. In fact, the pay of low-wage workers rose rapidly in this period.
Anyhow, if we don’t cherry pick our starting points, we can see that the real (inflation-adjusted) weekly wage was almost exactly the same in March of 2023 as it was in February of 2020, just before the pandemic hit.[1] But the average weekly wage is only part of the earnings story.
The unemployment rate in January of 2021 was 6.3 percent; it was 3.5 percent last month. This meant that many more people were working. The employment rate, the flip side of the unemployment rate, went from 57.5 percent of the adult population, when President Biden took office, to 60.4 percent in March.
This means that workers were 5.0 percent more likely to be employed in March of this year than in January of 2021. According to advanced economic theory, people who work earn more money than people who don’t. If we’re doing a serious comparison of people’s earnings, we have to consider the likelihood that they have a job. Speaker McCarthy’s numbers did not.
If McCarthy’s numbers on earnings are a bit screwy, the other part of his calculation is even worse. Kessler tells readers:
“Higher borrowing costs add another $1,500 to the figure, for a total of about $7,100. His calculation is based on higher mortgage rates on the median sales price of an existing home that have resulted as the Federal Reserve has jacked up interest rates in an effort to tame inflation.”
As Kessler points out, most people do not buy a home every month, so adding this $1,500 figure into the mix is rather dubious. But the actual story is even worse. While it has largely escaped the attention of business reporters, tens of millions of people refinanced their homes in the years from 2020 to 2022, before rates started to rise.
According to the Federal Reserve Board, nearly one-fourth of all homeowners (roughly 17 million households) refinanced their mortgages in 2021. A study using data from the prior year found average savings of $279 a month ($3,348 a year) for homeowners who refinanced their mortgage.
In other words, if we want to make serious comparisons of economic well-being, and factor in mortgage interest costs, this would be a big positive in the Biden years. Nearly 20 million homeowners (many refinanced in the first months of 2022) refinanced their mortgages, saving themselves thousands of dollars in interest payments.
In short, there is literally nothing to Speaker McCarthy’s story of families taking a big hit to their income under Biden. Real wages are back to their pre-pandemic level and people are far more likely to be employed than when he took office. And, tens of millions of households are saving thousands of dollars a year on mortgage interest payments.
McCarthy’s claim is pants on fire stuff; it doesn’t depend on the math.
[1] In fact, it was 0.2 percent lower in March of 2023, but the data do fluctuate. If Speaker McCarthy wants to make a big deal out of this 0.2 percent drop, it is probably worth mentioning that the March 2023 wage was 0.5 percent higher than the January 2020 wage, or that the January 2023 wage was 0.3 percent higher than the February 2020 wage.
Glenn Kessler, the Washington Post’s long-time fact checker, usually tries to be fair, but he really fell down on the job in a piece evaluating Kevin McCarthy’s claim that Biden’s policies cost families $7,400 a year. As the headline tells readers, Kessler concluded that it “depends on your math.” Nope, it depends on being honest.
As Kessler explained, the number came from E.J. Antoni, a research fellow in regional economics with the Heritage Foundation’s Center for Data Analysis. There are two parts to Antoni’s calculation.
The first part looks at average weekly earnings. Antoni said that average real weekly earnings were 5.1 percent lower in the most recent month than when President Biden took office in January of 2021. This loss came to $2,802.77, over the course of a year. Antoni doubles this figure, assuming two workers per household, which gives him a loss of $5,605.53 due to lower pay.
As Kessler sort of points out, this figure is distorted by composition effects. Millions of lower-paid workers lost their jobs in 2020, which raised the average wage of those still working. We can see that story clearly if we look at the longer picture.
As shown below, average real earnings soared in April of 2020, as millions of workers in hotels, restaurants, and other low-paying industries lost their jobs.
Source: Bureau of Labor Statistics and author’s calculations.
Pay fell as workers in lower-paying industries got their jobs back. This decline did not reflect a fall in the real wages of individual workers but rather a change in the mix of workers. In fact, the pay of low-wage workers rose rapidly in this period.
Anyhow, if we don’t cherry pick our starting points, we can see that the real (inflation-adjusted) weekly wage was almost exactly the same in March of 2023 as it was in February of 2020, just before the pandemic hit.[1] But the average weekly wage is only part of the earnings story.
The unemployment rate in January of 2021 was 6.3 percent; it was 3.5 percent last month. This meant that many more people were working. The employment rate, the flip side of the unemployment rate, went from 57.5 percent of the adult population, when President Biden took office, to 60.4 percent in March.
This means that workers were 5.0 percent more likely to be employed in March of this year than in January of 2021. According to advanced economic theory, people who work earn more money than people who don’t. If we’re doing a serious comparison of people’s earnings, we have to consider the likelihood that they have a job. Speaker McCarthy’s numbers did not.
If McCarthy’s numbers on earnings are a bit screwy, the other part of his calculation is even worse. Kessler tells readers:
“Higher borrowing costs add another $1,500 to the figure, for a total of about $7,100. His calculation is based on higher mortgage rates on the median sales price of an existing home that have resulted as the Federal Reserve has jacked up interest rates in an effort to tame inflation.”
As Kessler points out, most people do not buy a home every month, so adding this $1,500 figure into the mix is rather dubious. But the actual story is even worse. While it has largely escaped the attention of business reporters, tens of millions of people refinanced their homes in the years from 2020 to 2022, before rates started to rise.
According to the Federal Reserve Board, nearly one-fourth of all homeowners (roughly 17 million households) refinanced their mortgages in 2021. A study using data from the prior year found average savings of $279 a month ($3,348 a year) for homeowners who refinanced their mortgage.
In other words, if we want to make serious comparisons of economic well-being, and factor in mortgage interest costs, this would be a big positive in the Biden years. Nearly 20 million homeowners (many refinanced in the first months of 2022) refinanced their mortgages, saving themselves thousands of dollars in interest payments.
In short, there is literally nothing to Speaker McCarthy’s story of families taking a big hit to their income under Biden. Real wages are back to their pre-pandemic level and people are far more likely to be employed than when he took office. And, tens of millions of households are saving thousands of dollars a year on mortgage interest payments.
McCarthy’s claim is pants on fire stuff; it doesn’t depend on the math.
[1] In fact, it was 0.2 percent lower in March of 2023, but the data do fluctuate. If Speaker McCarthy wants to make a big deal out of this 0.2 percent drop, it is probably worth mentioning that the March 2023 wage was 0.5 percent higher than the January 2020 wage, or that the January 2023 wage was 0.3 percent higher than the February 2020 wage.
Read More Leer más Join the discussion Participa en la discusión
• InequalityLa DesigualdadUnited StatesEE. UU.Wall StreetEl Mundo Financiero
Okay, that may not be as dramatic as imagining world peace, but hey, I’m an economist. And, as I will argue, more reasonably paid CEOs would be a pretty big deal.
Just to set the table, CEOs have always been well-paid. At least in principle, it is a demanding job requiring skills in many areas. I said “in principle” because the corporate scandals of the last few decades have surfaced many examples of CEOs whose primary skill seems to be in the art of bullshitting. But there can be little doubt that maintaining a well-run company does require serious skills and hard work.
However, being a well-paid CEO means something very different today than it did fifty years ago. Back then, CEO pay was twenty to thirty times the pay of a typical worker. That would translate into roughly $2 million to $3 million a year given current pay structures.
Today, the average CEO of a major company is paid almost 400 times what an average worker earns, or roughly $25 million a year. There are issues of measurement that could make this figure higher or lower, but there is little doubt about the basic story where CEO pay has soared relative to the pay of ordinary workers and even relative to most other highly paid workers. [1]
While many of us might be offended by $25 million paychecks (the pay of 800 workers putting in a full year at the federal minimum wage) even if CEOs in some sense “earned” their pay, there is good reason to believe they don’t. There have been many studies showing that CEO pay does not closely correspond to returns to shareholders. A few examples are here, here, and here. Lucian Bebchuk and Jesse Fried’s book, Pay Without Performance, presents a wide range of evidence on this issue.
There is a simple story as to how CEO pay could become so divorced from their actual value to the companies they run. The basic picture is that there is no one to hold their pay down. We all understand how the pay of ordinary workers — retail clerks, assembly line workers, table servers in restaurant — are held down. Managers are supposed to keep their pay as low as possible, in order to maximize corporate profits.
If a company finds that they are paying their workers more than their competitors, they are likely to cut their pay to bring it into line. If current managers aren’t up to the task, the bosses find managers who will cut pay.[2]
But there is no remotely corresponding mechanism for CEO pay. The people who are supposed to hold their pay in check are the corporate boards of directors. The boards of directors ostensibly answer to shareholders, who elect the members of the board at regular intervals. However, as a practical matter, it is very difficult for shareholders to displace board members. More than 99 percent of the board members who are nominated by the rest of the board for re-election win their seats.
Being a board member is a very cushy job, typically paying well over $100,000 a year, and sometimes $300,000 or $400,000, for less than 400 hours of work, so board members generally want to keep their jobs. Since being re-nominated by the board virtually guarantees re-election, the best way to ensure that you get to hold onto your seat is to stay on good terms with other board members.[3]
This likely means not asking pesky questions like “can we pay our CEO less money?” Since top management plays a large role in selecting board members, it is not surprising that corporate boards tend to identify with them rather than shareholders. In fact, a recent survey of corporate directors found that the overwhelming majority did not even see limiting CEO pay as part of their job. Instead, they viewed their main role as serving the goals of management.
In this context, it is easy to tell a story where CEO pay can spiral upward almost without limit. Corporate boards are sitting on huge piles of money. They like their CEO and want to keep them happy. This means that they want to make sure their CEOs are paid at least as much as CEOs at their competitors. They do surveys so that they know what their competitors pay, and then add a ten or twenty percent premium.
And then their competitors do the same thing. They repeat this process every few years. In this story, CEO pay can only go up.
CEO Pay and Other High Earners
There has been much research in recent years showing how the pay of ordinary workers might bear little relationship to their actual output. There are two main stories. First, it appears that monopsonistic employment relationships are far more common than had previously been appreciated.
Rather than being an exceptional case, where for example there is one large employer in a small town, it seems many, if not most, employers have some degree of monopsony power. This means that rather than facing a horizontal supply curve, where they can hire as many workers as they want at the prevailing wage, paying one worker more money will also require paying other workers more money as well. In this context, workers will typically get paid less than their marginal product.
The other issue is that it seems workers may not typically be aware of their actual value in the labor market. They may not recognize how much they could earn at a different job, and therefore may accept pay that is less than their marginal product.
In both of these stories we can get situations persisting indefinitely where large numbers of workers may be earning substantially less than their marginal product. While this view is increasingly accepted for workers at the middle and bottom of the wage distribution, it is worth asking whether we may see comparable distortions at the high end, although in this case it would be a story with workers earning above their marginal product.
It is relatively easy to tell this story for CEOs and other top management, as discussed above. There basically is no market mechanism that would ensure CEO is kept in check. The corporate boards, that are supposed to work for shareholders in restraining in pay, tell us that they don’t even see this as part of their job. Of course, even if they said they were trying to rein in CEO pay, that is no guarantee they succeed, but when they tell us they are not even trying, we can be reasonably certain that they are not succeeding.[4]
But if pay for CEOs and other top managers is out of line with their marginal products, is it reasonable to believe that this is also likely the case for other highly paid workers. The story here is that the pay of CEOs and other top managers are a point of reference for other top-tier executives and high-level workers outside of the corporate sector.
We can think that workers enjoy a pay premium above their marginal product based on how close they are to corporate CEOs, both in the corporate hierarchy, and also in society more generally for people working outside the corporate sector. For other top-level executives this premium may be a large fraction of the CEO’s premium — say, something like 50 percent. This would mean that if the CEO gets $20 million more than their real value to the company, the chief financial officer and other C-suite executives get $10 million more than their real value.
For the next tier, the fraction might be something like 10 percent. In this case where the CEO is overpaid by $20 million, the third-tier executives will end up with $2 million more than their value to the company. There may still be some premium lower down, but once you get to the assembly line worker, it’s a safe bet it is zero.
To be clear, a third-tier executive is not getting inflated pay directly as a result of the inflated pay going to the CEO. Rather the CEO’s inflated pay is setting a pay structure (we use to talk about “wage contours”), that leads both workers and the people setting pay to think that the work of the third-tier executive is worth more than it actually is.
The same would apply to people working outside of the corporate sector. A university president may get something like an 8 percent CEO pay premium, which would mean that $20 million in excess pay for the average CEO is raising their pay by $1.6 million. The next level of university administration may get something like a 4 percent CEO premium, with excessive CEO pay adding $800,000 a year to their paychecks.
If this story is accurate, then it would not be easy to disrupt the pay structure. If a top university decided to pay its president $1 million rather than $2 million, it would be seen as an insult, given prevailing pay scales. An incumbent president would likely leave if they faced this sort of pay cut and many potential candidates for an open position offering half of the prevailing pay for university presidents would opt not to seek the job.
The norms around pay have real power in the world. If workers, and especially high-end workers, don’t feel they are being paid fairly, it is likely to show up in their performance. And, even if the pay for CEOs may exceed their value, a CEO or high-end worker who is trying to sabotage their employer is definitely in a situation to do considerable damage. This means it is difficult for an individual company to try to make the pay of their higher-level employees more closely reflect their actual value.
This suggests that bringing the exorbitant pay of CEOs, and other high-level workers, back in line with their value, will have to involve a process that takes place through time, similar to the process that led to the run-up in pay over the last five decades, but in the opposite direction. And, it should start at the top.
Setting the Ship Straight – Getting the Incentives Right
There have been various proposals for lowering the pay of CEOs, many of which involve some form of direct government intervention, such as a tax penalty for companies with overpaid CEOs. While corporations, with or without overpaid CEOs, can certainly afford to pay more taxes, I respect their enormous ability to avoid taxation. (Here’s my scheme for cracking down on corporate tax avoidance/evasion.)
I prefer a route that changes incentives. At it stands now, the corporate boards that most directly determine CEO pay have essentially zero incentive to try to lower pay. We should take seriously what these boards tell us; they see their jobs as serving top management. They are sitting on piles of money, which are not theirs. They know that they can keep top management happy by giving them generous paychecks. In this context, bloated CEO pay should not surprise us.
But we can change the incentives. The Dodd-Frank financial reform legislation included a “Say on Pay” provision, which required companies to send out their CEO pay package for shareholder approval at three-year intervals. The vote is simply a yes or no vote. There is no direct consequence for a pay package being voted down, but presumably it is an embarrassment to both the CEO and the board.
As it stands, very few pay packages are voted down. Less than 3.0 percent of Say on Pay votes lose. It is very difficult to organize diffuse shareholders, and since there is little consequence to a “no” vote, there is not much incentive for anyone to try.
However, we could put some meat on the bones. Suppose that the board of directors would lose their pay for the year if a CEO package was voted down. This would be a clear substantive outcome that might encourage more shareholders, who are disturbed by an excessive pay package, to make the effort to try organize among shareholders.
It is also likely that this would get the attention of corporate boards. My guess is that once two or three boards were forced to sacrifice their paychecks as a result of losing a Say on Pay vote, directors would become far more careful in dishing out dollars to top executives. In that world, they may decide it is actually a good thing if their CEO was paid somewhat less than their peer group. And, over time, we might see CEO pay fall back in line with their actual productivity.
While many people seem to view this plan to punish directors for excessive CEO pay as a left-wing proposal, it is hard to understand what is radical about giving shareholders more control over the company they ostensibly own. Presumably, this is exactly what fans of the free market would want.
After all, what possible incentive would shareholders have for paying a CEO less than their true value? This would mean that they would get lower returns on their stock if they ended up with an inept CEO because they weren’t paying the market price for a competent one.
Of course, it is hard to say how corporate boards would respond, and maybe CEOs really are worth their eight-figure paychecks, in spite of the evidence to the contrary. But it seems worth a try. We know that under the current system, there is no one who has the job of keeping CEO pay in check, and we pay a high price in the form of inequality, and less money for everyone else, as a result of the bloated pay structure at the top.
There seems little harm in trying to get the market to function as the textbooks say it does, and have the CEOs actually work for shareholders.
[1] The analysis cited, by Josh Bivens and Jori Kandra, excludes the pay of Tesla CEO Elon Musk. In 2021, the year analyzed, he cashed out stock options worth $23.5 billion. Had this been included in the sample of CEOs, it would have pushed average CEO pay for the sample to almost $100 million. On the other hand, Bivens and Kandra use the realized value of stock options rather than the value at the point where they are issued. Arguably, the latter is a better measure of compensation, since that is most immediately what the CEO is paid.
[2] There are exceptions, like Costco, who quite explicitly pay their workers more than competitors, but this is done with the idea that they will get more loyalty and more productive workers. This is a deliberate policy — it is not an accident.
[3] The board dynamics around CEO pay are discussed in Steven Clifford’s great book, the CEO Pay Machine.
[4] It is sometimes argued that the pay of CEOs at companies owned by private equity provides a good test of the true worth of CEOs, since there is no issue of diffused shareholder ownership. The pay of CEOs at private equity owned companies tends to be comparable or higher than their pay at publicly traded companies. However, the meaning of this comparison is questionable. Private equity firms usually look to hold a company for only a few years, working a major restructuring and then reselling it as a publicly traded company. This means both that they are likely making extraordinary demands on a CEO during this period, and that they face exceptional risk from bad performance. If they paid less than prevailing CEO salaries, they might get poor performing CEOs who would doom their project.
Okay, that may not be as dramatic as imagining world peace, but hey, I’m an economist. And, as I will argue, more reasonably paid CEOs would be a pretty big deal.
Just to set the table, CEOs have always been well-paid. At least in principle, it is a demanding job requiring skills in many areas. I said “in principle” because the corporate scandals of the last few decades have surfaced many examples of CEOs whose primary skill seems to be in the art of bullshitting. But there can be little doubt that maintaining a well-run company does require serious skills and hard work.
However, being a well-paid CEO means something very different today than it did fifty years ago. Back then, CEO pay was twenty to thirty times the pay of a typical worker. That would translate into roughly $2 million to $3 million a year given current pay structures.
Today, the average CEO of a major company is paid almost 400 times what an average worker earns, or roughly $25 million a year. There are issues of measurement that could make this figure higher or lower, but there is little doubt about the basic story where CEO pay has soared relative to the pay of ordinary workers and even relative to most other highly paid workers. [1]
While many of us might be offended by $25 million paychecks (the pay of 800 workers putting in a full year at the federal minimum wage) even if CEOs in some sense “earned” their pay, there is good reason to believe they don’t. There have been many studies showing that CEO pay does not closely correspond to returns to shareholders. A few examples are here, here, and here. Lucian Bebchuk and Jesse Fried’s book, Pay Without Performance, presents a wide range of evidence on this issue.
There is a simple story as to how CEO pay could become so divorced from their actual value to the companies they run. The basic picture is that there is no one to hold their pay down. We all understand how the pay of ordinary workers — retail clerks, assembly line workers, table servers in restaurant — are held down. Managers are supposed to keep their pay as low as possible, in order to maximize corporate profits.
If a company finds that they are paying their workers more than their competitors, they are likely to cut their pay to bring it into line. If current managers aren’t up to the task, the bosses find managers who will cut pay.[2]
But there is no remotely corresponding mechanism for CEO pay. The people who are supposed to hold their pay in check are the corporate boards of directors. The boards of directors ostensibly answer to shareholders, who elect the members of the board at regular intervals. However, as a practical matter, it is very difficult for shareholders to displace board members. More than 99 percent of the board members who are nominated by the rest of the board for re-election win their seats.
Being a board member is a very cushy job, typically paying well over $100,000 a year, and sometimes $300,000 or $400,000, for less than 400 hours of work, so board members generally want to keep their jobs. Since being re-nominated by the board virtually guarantees re-election, the best way to ensure that you get to hold onto your seat is to stay on good terms with other board members.[3]
This likely means not asking pesky questions like “can we pay our CEO less money?” Since top management plays a large role in selecting board members, it is not surprising that corporate boards tend to identify with them rather than shareholders. In fact, a recent survey of corporate directors found that the overwhelming majority did not even see limiting CEO pay as part of their job. Instead, they viewed their main role as serving the goals of management.
In this context, it is easy to tell a story where CEO pay can spiral upward almost without limit. Corporate boards are sitting on huge piles of money. They like their CEO and want to keep them happy. This means that they want to make sure their CEOs are paid at least as much as CEOs at their competitors. They do surveys so that they know what their competitors pay, and then add a ten or twenty percent premium.
And then their competitors do the same thing. They repeat this process every few years. In this story, CEO pay can only go up.
CEO Pay and Other High Earners
There has been much research in recent years showing how the pay of ordinary workers might bear little relationship to their actual output. There are two main stories. First, it appears that monopsonistic employment relationships are far more common than had previously been appreciated.
Rather than being an exceptional case, where for example there is one large employer in a small town, it seems many, if not most, employers have some degree of monopsony power. This means that rather than facing a horizontal supply curve, where they can hire as many workers as they want at the prevailing wage, paying one worker more money will also require paying other workers more money as well. In this context, workers will typically get paid less than their marginal product.
The other issue is that it seems workers may not typically be aware of their actual value in the labor market. They may not recognize how much they could earn at a different job, and therefore may accept pay that is less than their marginal product.
In both of these stories we can get situations persisting indefinitely where large numbers of workers may be earning substantially less than their marginal product. While this view is increasingly accepted for workers at the middle and bottom of the wage distribution, it is worth asking whether we may see comparable distortions at the high end, although in this case it would be a story with workers earning above their marginal product.
It is relatively easy to tell this story for CEOs and other top management, as discussed above. There basically is no market mechanism that would ensure CEO is kept in check. The corporate boards, that are supposed to work for shareholders in restraining in pay, tell us that they don’t even see this as part of their job. Of course, even if they said they were trying to rein in CEO pay, that is no guarantee they succeed, but when they tell us they are not even trying, we can be reasonably certain that they are not succeeding.[4]
But if pay for CEOs and other top managers is out of line with their marginal products, is it reasonable to believe that this is also likely the case for other highly paid workers. The story here is that the pay of CEOs and other top managers are a point of reference for other top-tier executives and high-level workers outside of the corporate sector.
We can think that workers enjoy a pay premium above their marginal product based on how close they are to corporate CEOs, both in the corporate hierarchy, and also in society more generally for people working outside the corporate sector. For other top-level executives this premium may be a large fraction of the CEO’s premium — say, something like 50 percent. This would mean that if the CEO gets $20 million more than their real value to the company, the chief financial officer and other C-suite executives get $10 million more than their real value.
For the next tier, the fraction might be something like 10 percent. In this case where the CEO is overpaid by $20 million, the third-tier executives will end up with $2 million more than their value to the company. There may still be some premium lower down, but once you get to the assembly line worker, it’s a safe bet it is zero.
To be clear, a third-tier executive is not getting inflated pay directly as a result of the inflated pay going to the CEO. Rather the CEO’s inflated pay is setting a pay structure (we use to talk about “wage contours”), that leads both workers and the people setting pay to think that the work of the third-tier executive is worth more than it actually is.
The same would apply to people working outside of the corporate sector. A university president may get something like an 8 percent CEO pay premium, which would mean that $20 million in excess pay for the average CEO is raising their pay by $1.6 million. The next level of university administration may get something like a 4 percent CEO premium, with excessive CEO pay adding $800,000 a year to their paychecks.
If this story is accurate, then it would not be easy to disrupt the pay structure. If a top university decided to pay its president $1 million rather than $2 million, it would be seen as an insult, given prevailing pay scales. An incumbent president would likely leave if they faced this sort of pay cut and many potential candidates for an open position offering half of the prevailing pay for university presidents would opt not to seek the job.
The norms around pay have real power in the world. If workers, and especially high-end workers, don’t feel they are being paid fairly, it is likely to show up in their performance. And, even if the pay for CEOs may exceed their value, a CEO or high-end worker who is trying to sabotage their employer is definitely in a situation to do considerable damage. This means it is difficult for an individual company to try to make the pay of their higher-level employees more closely reflect their actual value.
This suggests that bringing the exorbitant pay of CEOs, and other high-level workers, back in line with their value, will have to involve a process that takes place through time, similar to the process that led to the run-up in pay over the last five decades, but in the opposite direction. And, it should start at the top.
Setting the Ship Straight – Getting the Incentives Right
There have been various proposals for lowering the pay of CEOs, many of which involve some form of direct government intervention, such as a tax penalty for companies with overpaid CEOs. While corporations, with or without overpaid CEOs, can certainly afford to pay more taxes, I respect their enormous ability to avoid taxation. (Here’s my scheme for cracking down on corporate tax avoidance/evasion.)
I prefer a route that changes incentives. At it stands now, the corporate boards that most directly determine CEO pay have essentially zero incentive to try to lower pay. We should take seriously what these boards tell us; they see their jobs as serving top management. They are sitting on piles of money, which are not theirs. They know that they can keep top management happy by giving them generous paychecks. In this context, bloated CEO pay should not surprise us.
But we can change the incentives. The Dodd-Frank financial reform legislation included a “Say on Pay” provision, which required companies to send out their CEO pay package for shareholder approval at three-year intervals. The vote is simply a yes or no vote. There is no direct consequence for a pay package being voted down, but presumably it is an embarrassment to both the CEO and the board.
As it stands, very few pay packages are voted down. Less than 3.0 percent of Say on Pay votes lose. It is very difficult to organize diffuse shareholders, and since there is little consequence to a “no” vote, there is not much incentive for anyone to try.
However, we could put some meat on the bones. Suppose that the board of directors would lose their pay for the year if a CEO package was voted down. This would be a clear substantive outcome that might encourage more shareholders, who are disturbed by an excessive pay package, to make the effort to try organize among shareholders.
It is also likely that this would get the attention of corporate boards. My guess is that once two or three boards were forced to sacrifice their paychecks as a result of losing a Say on Pay vote, directors would become far more careful in dishing out dollars to top executives. In that world, they may decide it is actually a good thing if their CEO was paid somewhat less than their peer group. And, over time, we might see CEO pay fall back in line with their actual productivity.
While many people seem to view this plan to punish directors for excessive CEO pay as a left-wing proposal, it is hard to understand what is radical about giving shareholders more control over the company they ostensibly own. Presumably, this is exactly what fans of the free market would want.
After all, what possible incentive would shareholders have for paying a CEO less than their true value? This would mean that they would get lower returns on their stock if they ended up with an inept CEO because they weren’t paying the market price for a competent one.
Of course, it is hard to say how corporate boards would respond, and maybe CEOs really are worth their eight-figure paychecks, in spite of the evidence to the contrary. But it seems worth a try. We know that under the current system, there is no one who has the job of keeping CEO pay in check, and we pay a high price in the form of inequality, and less money for everyone else, as a result of the bloated pay structure at the top.
There seems little harm in trying to get the market to function as the textbooks say it does, and have the CEOs actually work for shareholders.
[1] The analysis cited, by Josh Bivens and Jori Kandra, excludes the pay of Tesla CEO Elon Musk. In 2021, the year analyzed, he cashed out stock options worth $23.5 billion. Had this been included in the sample of CEOs, it would have pushed average CEO pay for the sample to almost $100 million. On the other hand, Bivens and Kandra use the realized value of stock options rather than the value at the point where they are issued. Arguably, the latter is a better measure of compensation, since that is most immediately what the CEO is paid.
[2] There are exceptions, like Costco, who quite explicitly pay their workers more than competitors, but this is done with the idea that they will get more loyalty and more productive workers. This is a deliberate policy — it is not an accident.
[3] The board dynamics around CEO pay are discussed in Steven Clifford’s great book, the CEO Pay Machine.
[4] It is sometimes argued that the pay of CEOs at companies owned by private equity provides a good test of the true worth of CEOs, since there is no issue of diffused shareholder ownership. The pay of CEOs at private equity owned companies tends to be comparable or higher than their pay at publicly traded companies. However, the meaning of this comparison is questionable. Private equity firms usually look to hold a company for only a few years, working a major restructuring and then reselling it as a publicly traded company. This means both that they are likely making extraordinary demands on a CEO during this period, and that they face exceptional risk from bad performance. If they paid less than prevailing CEO salaries, they might get poor performing CEOs who would doom their project.
Read More Leer más Join the discussion Participa en la discusión
