April 18, 2019
Economists pride themselves on being the serious social science, the one most deserving of status as an actual science. I will let others make the comparative assessment, but there is an awful lot of nonsense that passes as serious analysis within economics. For cheap fun, I thought I would use a nice spring afternoon to highlight some of my favorites.
Myth 1: The Robots Are Taking All the Jobs
The “robots taking the jobs” story gets top place both because it is completely ridiculous and it is widely taken seriously in policy discussions. The story is ridiculous because it is directly contradicted by the data. Robots taking all the jobs is a story of rapid productivity growth. It means that we can produce the same output with fewer workers because the robots are doing work that used to be done by people. That is the definition of productivity growth.
But the productivity data refuse to cooperate with this story. This is not hard to discover. The Bureau of Labor Statistics releases data on productivity every quarter. The data show that productivity growth has been very weak in recent years, averaging just 1.3 percent annually since 2005. This compares to a rate of productivity growth of 3.0 percent in both the period from 1995 to 2005 and the long Golden Age from 1947 to 1973.
The job-killing robots story is sometimes diverted into a scenario that is just around the corner instead of being here today. Of course, we can’t definitely rule out that at some point in the future productivity will not accelerate sharply, but it is worth noting that this pickup does not seem to be on the immediate horizon. Investment is not especially high as a share of GDP, averaging 13.4 percent over the last three years. That compares to 14.2 percent from 1999 to 2001, and 14.4 percent from 1980 to 1982, its post-war peak.
Investment is not even especially high in the more narrow categories of computers and information technologies, where we would expect the robots to live. This means that if businesses are about to start a mass displacement of workers with robots, they don’t seem to be spending too much money on the process.
Perhaps the most important point here is that there is no reason to assume that a pickup in productivity growth would be associated with mass unemployment and a collapse of wage income. The 1947 to 1973 Golden Age was a period of rapid wage growth, as were the years from 1995 to 2001, when the stock bubble collapsed and the economy fell into recession.
Bad economic policy can prevent workers from sharing in the benefits of productivity growth (e.g. Peter Peterson-types demanding deficit reduction, thereby reducing output and employment), but it would be crazy to blame robots because Wall Street billionaires are slowing growth and pushing the economy toward recession.
Myth 2: We Are Threatened by a Declining Population
Our policy elite can be very flexible in their thinking. At the same time that many are worried that the robots are taking all the jobs, we also have the opposite threat frequently arising in public debates, that we are running out of workers. The basic story is that the baby boomers are retiring. The generation that followed is smaller. Furthermore, fertility rates among young people are falling to record lows. So, we now face the horrible problem that we are running out of workers just as the rise of robots means we are running out of jobs.
The running out of workers story also involves incredibly sloppy thinking. Yes, a decline in the workforce means that we will have fewer workers to support each retiree. But we have been seeing declining ratios of workers to retirees for the last eight decades. This stems from the problem that people are living longer.
The declining ratio of workers to retirees has not prevented both workers and retirees from seeing rises in living standards for the simple reason that the impact of productivity growth in raising living standards swamps the impact of demographics in lowering living standards. (It’s also worth noting that children have to be supported by workers as well, so the ratio of dependents to workers is actually considerably lower today than at its peak in 1965.)
The declining population story is sometimes turned into a fear of slower growth story. Other things equal, slower workforce growth will mean slower GDP growth. This should bring the obvious response, “who cares?” We should care about rising living standards, in which per capita growth is a factor. There is no obvious benefit in having more rapid growth due to a larger population. In fact, insofar as there are negative externalities not picked up in GDP (e.g. congestion and pollution), we should be happier if the population is growing less rapidly or shrinking.
There is a serious issue here that because of poor family supports, such as the availability of quality daycare and inadequate family leave policy, many people feel they cannot afford to have children. That is a genuine problem, but this is because people should be able to have children, not because society needs their kids.
Myth 3: Technology Is Shifting Income Upwards to the More Skilled
The slightly coherent version of the robots taking our jobs story is that technology is causing an upward redistribution of income to those who have the sophisticated skills needed to master the technology. This is the old “skills-biased technical change” argument that my friends Larry Mishel and Jared Bernstein did a great job decimating two decades ago.
Not only does the data not support the story of an increasingly rapid shift in labor demand to more skilled labor, the logic of the argument skips a key factor. The price commanded by new technologies, and therefore the demand for associated labor, is determined by policy, not the technology.
To be specific, the amount of money that goes to people developing software, artificial intelligence, pharmaceuticals, robots and other areas thought of as being at the technological frontier depends on the length and strength of patent and copyright protection. Shorter and weaker protections (or no protections) would send the price of these items plummeting. It would also mean much less demand for labor in these areas and much lower pay for the highly skilled workers in related occupations.
Arguably we benefit from having stronger and longer patent and copyright protection (I don’t think so), but the fact that these are government policies is not arguable. Insofar as there actually is a shift in income to people with skills in technology-related fields this is a result of policy choices, not the technology itself.
Myth 4: The Government Debt Will Impoverish Our Kids
An evergreen whine from the policy elite is that the government debt is going bankrupt our kids because they will be stuck paying it off. This one is so off the mark that it is difficult to know where to begin.
First of all, the debt involves payments within a generation, not between generations. At some point all of us profligate types will all be dead, so we will not be around to collect the interest on the debt. The rich ones among us will own government bonds, which they will presumably pass on to their kids. So the debt is then not an issue where our children and grandchildren will be paying us interest, but rather they will be paying interest to the children and grandchildren of Bill Gates and his friends.
That might not be a great story for those who are not children of the rich, but it can be fixed with that old-fashioned remedy of taxing the rich. In any case, the question is one of intra-generational distribution, not inter-generational distribution. It’s also worth noting that even with our relatively high debt-to-GDP ratio, the interest burden of the debt is less than 1.0 percent of GDP. That compares to the more than 3.0 percent of GDP that us baby boomer types had to pay in the 1990s when we were young.
The more serious version of the debt impoverishing our kids is that government deficits are pushing up interest rates, which in turn crowds out investment and slows growth. With nominal and real interest rates at extraordinarily low levels over the last decade, it is hard to tell this one with a straight face.
It is also worth noting that we have lost far more productive capacity (potential GDP) due to the austerity demanded by deficit hawks than we ever could have plausibly lost due to the crowding out story. In other words, the policies demanded by the “government debt will impoverish our kids” gang will cause our kids to have much lower real incomes. (The story is even worse when we consider that demands for austerity prevented measures to slow global warming.)
The other aspect missing from the debt story is that government debt is only one way that the government creates commitments for the future. When it grants patent and copyright monopolies it is also directing flows of income, often for decades in the future.
These monopolies are alternative mechanisms for paying for activities. Rather than directly paying for innovation and creative work with public funds, the government threatens to arrest people who violate the monopolies it grants, thereby allowing the monopoly holders to charge prices far above the free market price.
This gap is often quite large. By my calculations, government-granted monopolies cost us close to $370 billion (1.8 percent of GDP) a year in the case of prescription drugs alone and possibly as much as $1 trillion (5.0 percent of GDP) in total. Incredibly, the “government debt will impoverish our kids” gang never talks about the burdens created by patent and copyright monopolies.
Myth 5: Economy Threatening Bubbles Are Difficult to Detect
We saw the tenth anniversary of the Lehman bankruptcy last fall and with it a whole round of papers and conferences dedicated to the problems of the financial crisis. There was virtually no discussion of the housing bubble, the collapse of which was the real cause of the Great Recession.
There is a huge sense in which the focus on financial crises is self-serving for the economics profession. Financial instability can be difficult to anticipate. After all, many financial assets are complex, with new assets constantly being created. In many cases, there is limited data that is publicly available. Who knew that AIG had issued $600 billion worth of credit default swaps against mortgage-backed securities? This gives economists an excuse for missing the Great Recession.
By contrast, the housing bubble, whose crash sank the economy, was easy to see for anyone who pays attention to economic data. We get monthly data from the Commerce Department on residential construction. We also have the quarterly GDP data in which residential construction is a major component. It required some determined ignorance for macroeconomists not to notice the huge jump in this category, which peaked at 6.7 percent of GDP in 2005. That compares to a normal level in the neighborhood of 4.0 percent of GDP.
The unprecedented run-up in house prices was also easy to see with publicly available data. After largely tracking the overall inflation rate in the post-war era, real house prices rose 70 percent from 1996 to 2006. The fact that this increase was associated with virtually no rise in real rents might have seen peculiar to people who know economics. Also, the fact that the vacancy rate was high and rising might also have suggested that this rise in house prices was not being driven by the fundamentals of the markets.
And, economists also should have known that a collapse of the housing bubble would lead to an end to the consumption driven by the ephemeral wealth created by the bubble. Some guy named Alan Greenspan wrote several papers on this consumption driven by housing wealth.
But rather than own up to having missed the obvious, the economics profession has decided that the problem was a mysterious financial crisis that caught everyone by surprise. Who knows where the next one may lurk?
As a policy matter, this actually has the positive effect of generating more political support for limiting abuses in the financial industry, since we are supposed to be worried that somehow bad practices will, by themselves, lead to another financial crisis and Great Recession. This is good, because finance is an intermediate good, like trucking.
We want the financial sector to be as small and simple as possible, there is no more reason to want a large financial sector than there is to want a trucking industry where goods are shipped back and forth for no reason. The financial industry has been both a major source of waste and inequality over the last four decades.
It would be best if we could rein in the financial sector based on an honest discussion of its role in the economy, but in American politics, no one expects honest discussions. So, we’ll settle for some positive spillover from economists’ efforts to cover their asses.
Myth 6: CEOs Manipulate Share Prices with Buybacks to Maximize Shareholder Value
One of the bizarre stories that passes for wisdom is that CEOs have gotten in the habit of committing large amounts of money to share buybacks in order to maximize shareholder value. Just about every part of this story is hard to square with reality.
First, if the goal is maximizing shareholder value, they are not doing a very good job. Stock returns have been extraordinarily low over the last two decades, averaging less than 5.0 percent annually, after adjusting for inflation, compared to a long-term average of 7.0 percent. It does seem at least a bit odd that when management is supposed to be exceptionally focused on providing returns to shareholders that they are doing such a poor job of it.
Another aspect of this story is that the reason that companies are not investing more is supposed to be that they are paying out so much money to shareholders. Looking at the data, what seems most striking is that there is not much fluctuation, apart from cyclical ups and downs in the share of GDP going to investment.
If investment is determined primarily by investment opportunities, and these tend not to change much, at least for the economy as a whole, then in a period of high profits like the present one, lots of money will be paid out to shareholders, since companies have nothing else to do with it. There is good reason to be unhappy with the rise of corporate profits at the expense of wages, but if companies don’t have good investment opportunities, it’s not obviously bad that they pay the money out to shareholders.
Finally, the focus on buybacks as being especially pernicious seems odd. Is there a reason anyone would be happier if the money was paid out as dividends instead? The switchover to buybacks instead of dividends as the main route for giving money to shareholders, dates from an early 1980s court ruling that approved the practice.
I have heard the complaint that buybacks are worse because top management can manipulate the timing in order to maximize the value of their stock options or planned sales. While that may true, they can also manipulate dividend announcements or statements on earnings.
More importantly, if we think management is manipulating the timing of buybacks, then the victims of the manipulation are shareholders who aren’t in on the joke. That is entirely possible, but then we are not telling a story of maximizing shareholder value. We are telling a story where CEOs and other top management are enriching themselves at the expense of shareholders.
That sets up a dynamic in which shareholders should be allies in cracking down on excessive CEO pay, but I’ll leave that one for another day.
Comments