Beat the Press is Dean Baker's commentary on economic reporting. He is a Senior Economist at the Center for Economic and Policy Research (CEPR). To never miss a post, subscribe to a weekly email roundup of Beat the Press.

Please also consider supporting the blog on Patreon.

Follow on Twitter Like on Facebook Subscribe by E-mail RSS Feed

Last week Joe Stiglitz wrote a piece that included criticisms of the Obama administration for having an inadequate stimulus in response to the Great Recession. Larry Summers, who had been head of Obama’s National Economic Council responded by defending the administration. Stiglitz had a follow-up response to Summers.

Both Stiglitz and Summers are very capable of speaking for themselves so I won’t try to summarize their arguments, but I do want to take issue with one point in Summers’ piece. In his response to Stiglitz, at one point Summers comments:

“He was a signatory to a November 19, 2008 letter also signed by noted progressives James K. Galbraith, Dean Baker, and Larry Mishel calling for a stimulus of $300–$400 billion — less than half of what the Obama administration proposed. So matters were less clear in prospect than in retrospect."

This is a serious misrepresentation of the letter, which I had helped to draft and circulate. This letter was very clearly referring to a one-year stimulus. It urged leaders in Congress to quickly pass a stimulus:

“The potential severity of the downturn suggests that a boost to demand on the order of 2.0-3.0 percent of GDP ($300-$400 billion) would be appropriate, with the goal being to get this money spent quickly.”

Comparing the sums in a one-year stimulus to the multi-year stimulus request put forward by the Obama administration is comparing apples to oranges.

Furthermore, this letter had been drafted in early October, before we knew the full severity of the downturn following the collapse of the Lehman. At the time of the drafting, we were looking at a job loss of 159,000 reported for September. At the point where the Obama administration was crafting its stimulus package in January of 2009, it was looking at job losses of more than 1.5 million over the prior three months. And the September job loss figure had been revised upward to more than 400,000.

To misrepresent our letter to justify the size of the Obama administration’s stimulus request involves, at the least, an extremely sloppy reading of a one-page document.

Add a comment

The NYT has a strange piece which treats the idea of a four day work week as a sort of alien concept that perhaps we will see in the 22nd century. It is bizarre because there has been a consistent shortening of the length of work years for well over a century. In Germany, France, and other northern European countries the average work year is between 1400 and 1500 hours. This is due to the fact that workers are now guaranteed 5 to 6 weeks a year of vacation, in addition to paid sick days and paid family leave. The United States has been to a large extent an outlier in that it has seen relatively little reduction in work time over the last four decades.

Presumably, this long trend towards shorter work years will continue. While it may not take the form of a four day work week, workers in many countries have already seen an equivalent reduction in work hours. It is also worth noting that if it really turns out that robots will eliminate many of the jobs that now exist (the productivity data point in the opposite direction) then shorter work years is an obvious way to keep people employed.

Add a comment

That was one of the items on the list of factors adding to inflationary pressures in China in this NYT article, but it doesn't seem very plausible. China is on track to import a bit less than $140 billion worth of goods and services from the United States this year. This is less than 1.0 percent of China's GDP measured at its exchange rate value. (Using a purchasing power parity measure this would be about 0.6 percent.)

Suppose that China's trade war tariffs add 30 percent to the price of these imports (a very high assumption). This would push up the overall price level by 0.3 percentage points. However, we are continually reminded that much of China's exports are primarily foreign value-added. Let's assume that 40 percent of US exports to China end up as value-added in exports either to the US or third countries. This means that the 30 percent rise in price due to tariffs would add 0.18 percentage points to the price level. That is not exactly soaring inflation.

The fact that the yuan has fallen about 6.0 percent against the dollar over the last six months will almost certainly have more impact on China's inflation rate. The decline in the value of the yuan was not mentioned in the piece.

Add a comment

Ten years ago we saw the culmination of a period of ungodly economic mismanagement with the collapse of Lehman Brothers and a full-fledged financial crisis. The folks who led us into this disaster rushed to do triage and tend to the most important problem: saving the bankrupt banks.

They also had to cover their tracks. They insisted that the financial crisis was some sort of fluke event — a lot of bad things went wrong simultaneously — and who could have predicted or prevented that? They had a lot of assistance in this coverup because almost all the people who did and wrote about economics at the time also missed the housing bubble and the harm that its inevitable collapse would cause.

The coverup continues to the present, largely because the same people who messed up in the years leading up to the crash are still in positions of authority. They are still the ones writing and talking about economics in major news outlets. So we can expect a lot of "who could have known?" drivel in the weeks ahead.

CEPR will be putting out a paper soon showing once again how the bubble was easy to recognize as was the fact that its collapse would be a disaster. Today I will just share one chart that shows much of the story.

The bubble led to an unprecedented run-up in house prices (with no accompanying rise in real rents), which in turn led to residential construction hitting 6.5 percent of GDP, more than two full percentage points above the long-term average. (But hey, who could have noticed that?)

In addition, the bubble-driven increase in housing wealth led to an unprecedented consumption boom as people spent based on their housing wealth. (This is called the "housing wealth effect" which was very old news by the time I was in graduate school in the 1980s.) This consumption boom could be seen in the plunge in the savings rate which is reported monthly by the Commerce Department. That fell to a low of 2.2 percent in 2005. That compares to an average in the years before the stock wealth effect drove down the savings rate in the 1990s of more than 7.0 percent. It is currently also hovering near 7.0 percent. (FWIW, savings data are subject to large revisions. At the time, the savings rate in 2005 was reported as -0.4 percent [Table 10]).

The question everyone should ask the "who could have known?" crowd is how could you miss the unprecedented run-up in house prices. Even more importantly, how did you miss the extent to which it was driving the economy through the construction and consumption boom? And finally, what on earth did you think would replace more than 5.0 percentage points of GDP worth of lost demand ($1,000 billion in today's economy) when this housing bubble burst?

Those are pretty simple questions, but you won't see people asking them in major news outlets. They have too much stake in maintaining the myth that people managing the economy really know what they are doing and the crash and financial crisis were fluke events that could not be foreseen. It ain't so.

Add a comment

In an NYT piece assessing the state of the labor market after Friday jobs report, Neil Irwin notes that the employment rate for prime age workers (ages 25 to 54) stood at 79.3 percent in August. That is the same as the 79.3 percent rate in February, indicating that there had been no increase over a six-month period.

However, this may be less compelling as an argument that the labor force is hitting its limits than it initially seems. In August of 2017, the employment rate (EPOP) for prime-age workers stood at 78.4 percent, which also turns out to be the same rate as in February of 2017. In August of 2016, it stood at 77.8 percent, which was also its level in February of 2016. And, in August of 2015, it stood at 77.2 percent. The EPOP for prime-age workers in February of 2015 was also 77.2 percent.

Here's the picture for the last four years.

Prime Age Employment to Population Ratio

prime age EPOP

Source: Bureau of Labor Statistics

The data are seasonally adjusted, so the fact that EPOPs have a habit of being the same in August as February can't be blamed on seasonal factors. It is simply a weird coincidence. But, this should be a warning against putting much stock in the view that the economy is hitting its limits based on the August employment data.

As the good book says, the household data are highly erratic. When you see a weird movement in one month's data, it is most likely a fluke. You should want to see two or three months before believing it is actually telling us something about the world.

Add a comment

NYT readers were no doubt disturbed to see a column in which former Fed Reserve Board chair Ben Bernanke, Obama Treasury Secretary Timothy Geithner, and Bush Treasury Secretary Henry Paulson patted themselves on the back for their performance in the financial crisis. First, as they acknowledge in the piece, all three completely failed to see the crisis coming.

During the years when house prices were getting way out of line with both their long-term trend and rents, Bernanke was a Fed governor, then head of the Council of Economic Advisers, and then Fed chair. He openly dismissed the idea that the run-up in house prices could pose any threat to the economy. Henry Paulson was at Goldman Sachs until he became Treasury Secretary in the middle of 2006. As the bank's CEO, he was personally profiting from the bubble as the bank played a central role in securitizing mortgage-backed securities. Timothy Geithner was president of the New York Fed, where he was paid over $400,000 a year to make sure that the Wall Street banks were not taking on excessive risk.

It is bad enough that these three didn't see the crisis coming, but they still seem utterly clueless. They tell readers:

"Productivity growth was slowing, wages were stagnating, and the share of Americans who were working was shrinking. That put pressure on family incomes even as inequality rose and upward social mobility declined. The desire to maintain relative living standards no doubt contributed to a surge in household borrowing before the crisis."

Actually, productivity growth didn't begin to slow until 2006, as the bubble was hitting its peak. Growth was quite strong from 2000 to 2005, which means the cause of wage stagnation in those years must lie elsewhere. (If they had access to government trade data they might think the explosion of the trade deficit to 6.0 percent of GDP played a role.) The surge in borrowing clearly preceded the productivity slowdown as could be seen from the plunge in savings rates or reading the papers celebrating people pulling equity out of their homes by some guy named Alan Greenspan.

Add a comment

The New York Times had a piece on how education secretary Betty DeVos is trying to block state efforts to prevent abusive practices by student loan servicers after she had weakened federal protections. Her argument is the same one that the federal government used to weaken state rules on mortgage lending practices during the housing bubble years — that federal law preempts state law.

I'll leave it to the lawyers to decide the legal question here, but the economic one is straightforward. If servicers can make money from abusive practices (e.g. harassing phone calls, illegal threats, and charging arbitrary fees), they will. That is what we expect in a free market economy, businesses try to maximize profit.

While the piece treats this as a consumer protection issue, which it is, it is also one of economic efficiency. If it is possible to make lots of money by ripping off students in servicing their loans, then businesses will devote resources to ripping off students rather than something productive.

Think of it like making stealing cars legal. If people could make lots of money by stealing cars, many will quit their day job and spend their time stealing other people's cars.

We should also understand the issue of government-issued or government-insured student loans for tuition at for-profit universities the same way. Many of these universities provide little in the way of education and do not give students a marketable skill. As a result, the default rate on loans at many of these schools is often over 40 or 50 percent.

For these for-profit colleges, the students are simply intermediaries to access to government money. They allow students to make tuition payments which they could not possibly do otherwise. The students basically get nothing for their money, but in a world where the government requires no accountability from the schools, this doesn't matter.

It seems that for Betty DeVos, the point of the student loan program is to make the people who own these for-profit colleges richer with taxpayer money. In this respect, it is probably worth noting that she made the dean of DeVry University, one of the biggest for-profits with a very high default rate, head of the student loan fraud division at the Education Department.

Add a comment

(This post originally appeared on my Patreon page.)

The Fact Check gang has been having a field day going after Bernie Sanders, Alexandria Ocasio-Cortez and other proponents of Medicare for All. The latest battle is over a study produced by the right-wing Mercatus which showed that a government-run health care program could reduce national health expenditures by $2 trillion over the course of a decade (roughly 0.8 percent of GDP).

Sanders and Ocasio-Cortez seized on this projection of savings coming out of a right-wing think tank as support for the greater efficiency of a universal Medicare program. Of course, this was not the point that the Mercatus folks intended people to get from their study. They highlighted the fact that their projections showed Medicare for All increasing costs to the federal government by $32.6 trillion over the first ten years of operation, with the amount equal to 10.7 percent of GDP in 2022 and rising to 12.7 percent of GDP in 2031.

The fact check crew definitely went overboard in attacking Sanders and Ocasio-Cortez for misrepresenting the study. One scenario in the study did, in fact, show that Medicare for All would reduce national health care expenditures by $2 trillion over the decade.

Add a comment

Robert Samuelson used his Labor Day column to tell us that our pay really didn't end up in the pockets of rich people. The problem is that it all went to employer-provided health care insurance. The argument is that health care costs have vastly exceeded the overall rate of inflation. Since a standard health care benefit is larger as a share of the pay of a low-wage worker than a high-wage worker, the increased cost of the benefit took away the money that otherwise would have gone into pay increases. He cites a survey (but doesn't link to it) that purports to show this.

The problem with this story is that it contradicts the data from the Bureau of Labor Statistics which show little change in the share of labor compensation going to employer-provided health insurance. This is true even in lower paying occupations.

For example, the share of compensation going for health benefits for workers in Production, transportation, and material moving occupations went from 8.5 percent in 2004 to 9.8 percent in 2018, according to the Bureau of Labor Statistics Employer Cost for Employee Compensation series. That means that if health insurance costs had remained at a constant share over this period, wages would be approximately 1.4 percentage points higher, adding 0.1 percentage point annually to wage growth. The series actually peaked at 10.4 percent in 2014, which means that declining health care costs should have been adding to wage growth for these workers in the last four years.

There is a similar story for office and administrative support occupations where the wage and salary share of compensation fell from 71.0 percent in 2004 to 69.3 percent in 2018. The latter figure is up from a low of 68.8 percent in 2014. Again, the declining wage share of compensation only explains a small part of the wage stagnation story.

There are three things going on here. First, lower-paid employees are much less likely to have health insurance coverage at their job than was the case two decades ago. Second, they are likely to have less generous coverage, with more deductibles and co-pays. Also, they more often have to pay part of the premium. Finally, in recent years, health care costs have largely moved in step with overall economic growth, which explains their declining share of compensation.

The Washington Post may always have room for people denying that there has been an upward redistribution of income, but it happens not to be true. There has been and it is enormous.

Add a comment

The other 49 states have a massive trade deficit with Michigan in cars because of the terrible trade deals that our stupid trade negotiators signed with Michigan. Thankfully, Trump is going to impose a 25 percent tariff on cars from Michigan going into the other 49 states to set things right.

That is pretty much how we should understand Trump's complaint about the trade surplus that Canada has with the United States. (Yes, the United States actually runs a trade deficit with Canada, when properly measured, even including services.) Canada has a trade surplus for pretty much the same reason that Michigan has a trade surplus in cars. It has historically set itself up as a good place to manufacture goods.

Note that Canada's trick is not low-cost labor. Its workers get comparable wages to workers in the US and they enjoy considerably more labor rights than do workers here. That is the same story with Michigan's auto industry where workers are more likely to be unionized and get somewhat higher wages on average than workers in the other 49 states. (There can be currency issues with Canada, but we'll skip that for now.)

Suppose we did put 25 percent tariffs on cars going from Michigan to the other 49 states. Would that mean more jobs in the auto industry in the other 49 states?

The answer to that is not clear. To some extent, the auto manufacturers that have operations in Michigan may just keep their factories going and split the tariff with their customers. This will mean fewer Michigan cars will be sold in the other 49 states. Since Michigan cars include many parts from the other 49 states, that will mean fewer jobs in the auto industry in the other 49 states.

Add a comment