Beat the Press is Dean Baker's commentary on economic reporting. Dean Baker is co-director of the Center for Economic and Policy Research (CEPR).

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

The big difference between outsourcing and robots is that the former is happening and the latter isn't. Productivity growth (a.k.a. robots) has been very slow in recent years. It has averaged less than 1.0 percent over the last seven years and has sometimes been negative.


By contrast, many firms are looking to outsource jobs, both domestically and internationally, on an ongoing basis. For this reason, when the NYT told readers in a story on the jobs report and the economy:

"Perhaps even more than outsourcing, the real threat to job growth for Mr. Trump’s blue-collar base comes from automation and other efforts to improve productivity on the factory floor."

It had the picture backward. At least for the immediate future, it does not seem rapid productivity growth will be a major source of job loss.

Add a comment

Landon Thomas Jr. had an NYT piece noting the peculiar divergence between the stock market, which has risen sharply since Donald Trump's election and the dollar, which has fallen. The article claims this is peculiar since both tend to move in the same direction, rising in a strong economy and falling in a weak economy.

Actually, this is not really true. There have been many long periods where they have gone in opposite directions. For example, the dollar peaked in the mid-80s and then fell through the rest of the decade. The stock market did crash in the fall of 1987 but then rose through the rest of the decade. The dollar fell against most currencies from 2001 to 2007 even as the market recovered from its crash beginning in the summer of 2002.

A weaker dollar can be good news for U.S. corporate profits since it means that domestically produced goods and services become relatively more competitive internationally. This could be a reason the two would move in opposite directions.

However, there is another story in this case which could plausibly explain the divergence. President Trump and the Republicans have made reducing corporate income taxes a priority. Trump has proposed outlandish treatments of pass-through corporations, which would allow them to pay just 15 percent on their income.

This is a total joke proposition: no serious economist thinks this is a way to treat these companies. It essentially allows every rich person in the country to pay a 15 percent tax rate on the bulk of their income, as opposed to the 25 percent rate currently paid by teachers and fire fighters and other middle-class workers. Almost none of them are so stupid that they can't figure out how to have their income show up in a pass-through corporation and the ones that are too stupid have accountants that can figure out how to tie their own shoes.

Add a comment

Eduardo Porter had an interesting piece discussing the prospects for tax reform in the NYT. While the piece correctly highlights some of the absurdities of the U.S. tax system, it may have given readers a wrong impression in some areas.

For example, it notes that income-related taxes are a far higher share of the tax burden in the U.S. than in other wealthy countries. It argues that this is bad because income taxes tend to be more of a drag on growth than taxes on consumption.

While there is clearly some truth to this, it is important to note that income taxes are far more progressive than consumption taxes. In other countries, where consumption taxes are higher, the government provides much more generous benefits to the public, such as national health care, more generous pensions, and free or low-cost college education. While it is possible that the public would support regressive taxes that support programs with broadly based benefits, as they do with Social Security, it is unlikely that they would support an increase in regressive taxes that are not tied to an expansion of benefits.

The piece also exaggerates the harm caused by the current tax system when it refers to the profits that corporations keep abroad to avoid paying taxes. There is little reason to believe that companies would invest more in the United States if they claimed these profits here. Corporations are currently flooded with cash, paying out large dividends and doing massive share buybacks. A lack of capital is not a major factor limiting most corporations investment.

Add a comment

The NYT had an article on Amazon's job fairs which were set up to recruit workers for 50,000 new jobs nationwide. At the end of the piece, the article discusses concerns that robots may soon replace the jobs that Amazon is now hiring for in its warehouses:

"Amazon is more aggressively using robots to help make the operations inside its warehouses more efficient. For now, the company said machines are not replacing people. Instead, they mostly move large shelves of merchandise to stations where orders are manually picked.

"Many academic researchers and start-ups are working on robots that have the dexterity to pick orders automatically. Amazon sponsors a competition to encourage engineers to build more advanced warehouse robots.

"When those technologies are perfected, the employment picture inside Amazon’s warehouses could look very different. That day could be a decade or more away, though."

It is important to remember that productivity growth has been at record low levels in the last five years, meaning that we are seeing very few workers displaced by robots. Furthermore, the Federal Reserve Board has been raising interest rates over the last year and a half because it is concerned the economy is creating too many jobs. The concern about budget deficits is also a concern about inadequate productivity growth (too much demand and not enough supply). 

In other words, in almost every other economic debate our concern is the opposite of having robots replacing workers. The concern is that we won't have enough goods and services to go around.

If robots create a distributional issue, that is because of policies like patent monopolies that give all the money to owners of robots. These policies can be changed, but not if the media has a policy of never talking about them.

Add a comment

It's popular among economists and policy types to wisely note that technology is leading the rich to get richer. Many of them consider this unfortunate, but hey, should we be Luddites and try to stop technology?

This is, of course, silly propaganda, but it passed for sophisticated thinking in policy circles. It is not technology, but our policy around it, like patent and copyright protection, that redistributes income upward. We got yet another lesson along these lines in an NYT article reporting that the Trump administration is beginning a major investigation on China's trade practices which will focus on its treatment of U.S. patents, copyrights, and other forms of intellectual property (IP). The implication is that we would impose retaliatory measures because China was hurting Bill Gates, Elon Musk, and other major beneficiaries of these government-granted monopolies in the United States.

The decision to focus on IP is striking since there is little dispute at this point that China's decision to deliberately keep down the value of its currency in the last decade badly hurt U.S. manufacturing. The result was the loss of millions of manufacturing jobs. This ruined the lives of many of these workers and devastated communities in places like Ohio and Pennsylvania. It also put downward pressure on the wages of non-college educated workers throughout the economy.

Furthermore, the trade deficit that resulted from China's currency practices is the main reason that the United States suffers from secular stagnation (a.k.a. inadequate demand). This is the reason growth was slow following the collapse of the housing bubble and even today, almost ten years after the start of the recession, we are still not back to full employment.

Anyhow, the plight of the bulk of the country's workers was apparently not a sufficient reason to get upset over China's trade policy. But not honoring Bill Gates' copyrights? That's serious stuff. And the folks who write and talk about economics will tell us it is just technology.

Add a comment

Given the hostility that President Trump and his followers have directed towards the media, several people have suggested a name change for my blog. While I understand and sympathize with the idea of not promoting violence toward the media, I don’t think BTP has contributed to this sort of hostility.

First, there are different meanings of the word “beat,” and I did intend to play off these differences in choosing the name. There is “beat” as in the sense of the Chicago Cubs beat the Cleveland Indians in the World Series. I like to think that in many areas I do a better job of discussing economic issues than most of the media.

For example, I have endlessly harangued reporters about writing large numbers, most importantly budget numbers, without any context. When people hear that the government is spending $20 billion on TANF or $30 billion on foreign aid, they think these are sizable sums. After all, none of us will ever see anything like this amount in our lifetime.

However, as a share of the federal budget these programs are pocket change, with the $20 billion for TANF being roughly 0.5 percent of total spending and $30 billion for foreign aid a bit less than 0.8 percent. Polls consistently show that people hugely over-estimate the share of the budget that goes to foreign aid, TANF, and other anti-poverty programs.

I know that many people want to believe that all their tax dollars go to foreign aid and poor people’s programs because they are racists who hate the people they think of as beneficiaries of these programs. But many of the people who think large shares of the budget go to these programs are not racist, they just hear $20 billion or $30 billion and think that is a lot of money.

It would be a very simple matter if reporters got in the habit of reporting these numbers in some context. Some people might still insist that all of our tax dollars go to TANF even if they constantly heard that it was 0.5 percent of the budget, but my guess is the public would be much better educated.

I consider it one of my BTP victories that I got then NYT Public Editor Margaret Sullivan to agree with me (with assists from Just Foreign Policy and Media Matters). I thought this would lead to a change in practice at the country’s leading newspaper, but unfortunately not. The big numbers still routinely appear without any context.

There have been a number of other areas where I think my commentary beats the major news outlets in economic reporting. I should say that I think economic reporting has improved considerably in the more than two decades that I have been commenting on it. I’d like to think that my calling attention to some seriously bad practices has played a role.

Add a comment

Greg Mankiw had a short NYT piece outlining the problems in providing health care. While some of what he said was reasonable, he ended with the tired cliche:

"The best way to navigate the problems of the health care marketplace is hotly debated. The political left wants a stronger government role, and the political right wants regulation to be less heavy-handed."

This is not at all true. The right tends to want stronger and longer patent and related protections for prescription drugs and medical equipment. These government-granted monopolies can raise prices by several thousand percent above the free market price. As any economist would expect, these monopolies create enormous problems of enforcement and lead to a wide variety of rent-seeking behavior, such as drug companies lying about the safety and effectiveness of their drugs in order to sell more of them.

While longer and stronger patent protection does redistribute income upward, it can hardly be described as "less heavy-handed" in a world where the government pays for research upfront and then allows drugs and medical devices to be sold in a free market. (Discussion of alternatives are here and here.)

Add a comment

Since several people have asked me why I bothered to write about Amazon's stock price possibly being overvalued, let me get out the sledge hammer and hit people over the head with the issue.

Let's imagine that this week the god of Wall Street comes down and announces that Amazon's proper valuation is half of its current level. (This is a hypothetical, not my target price for the stock.) Since the god of Wall Street is never wrong, we would expect that Amazon's stock price would quickly plunge, eliminating $240 billion in market value.

Has this information destroyed any wealth? Well, the folks who owned Amazon stock have $240 billion less than they did previously, so they clearly have less wealth. But the god of Wall Street knows the true value of Amazon stock, so there really was not any basis for this $240 billion in wealth. In effect, they would have this wealth and be able to spend based on it, as long as other investors did not realize that Amazon's stock was hugely over-valued.

In this sense, it can be seen as very similar to counterfeit money. Suppose Amazon stock was priced in line with the god of Wall Street's assessment, but Amazon shareholders collectively held $240 billion in counterfeit money that everyone accepted as though it was real. If the government suddenly discovered a way to detect these counterfeits so no one would ever be able to spend this money again, it would be the same story as the Amazon stock losing half its price.

The moral of the story is that there is no reason for those of us who don't hold large amounts of Amazon stock to be happy about it being overpriced if this is the case. It is the same story as people having large amounts of counterfeit money. It's good for them, but when they price the rest of us out of the housing market and their spending causes the Fed to raise interest rates and slow growth, it's not good for everyone else.

Add a comment

In their NYT piece on the possibilities for people switching jobs mid-career, Clair Cain Miller and Quoctrung Bui link to a piece by M.I.T. economist David Autor to support the assertion that extensive research shows middle skills jobs are disappearing. Actually, more careful research showed the opposite. In the last decade, both middle- and high-skills jobs (using Autor's definition) were declining as a share of total employment. Only the least skilled jobs had an increasing share.

It is also worth noting that there is little evidence for the "skills shortage" discussed in this piece. While businesses like to complain about not being able to get qualified workers, the ordinary response to a shortage is a rise in price. In other words, if businesses really were seeing a skills shortage, we would expect to see rapidly rising wages for significant groups of workers. We don't, which suggests that businesses are just whining because they think it will help them get something from the government — not because they actually can't get qualified workers.

Add a comment

The NYT had a piece touting the recent run up in Amazon's stock which briefly made Jeff Bezos the world's richest person. It then turns to Hendrik Bessembinder, a finance professor at Arizona State University, who describes the company as "one of the greatest wealth creators since 1926."

This designation as a "wealth creator" is based on its market capitalization of almost $500 billion. While this is a huge amount of money, it is not clear that Amazon's current or likely future profits justify this price. Ultimately, stockholders value a company based on the profits it makes for shareholders and its current profits nowhere near justify its $500 billion market capitalization.

There have been other companies in the recent past that had stock prices that bore no relationship to their profits. AOL and Priceline are two that stand out, both with market capitalizations of well over $100 billion at the peak of the stock bubble in 2000. In both cases, the shareholders in these companies would have seen a huge amount of wealth destroyed if they bought them near their peak price.

The question is whether Amazon will be able to join the elite group of wealth destroyers if its stock price falls to a level more in line with its profits.

Add a comment

A Washington Post article discussed the future prospects for the Affordable Care Act now that the repeal efforts seem to be defeated. At one point, the article referred to the cost-sharing subsidies in the program, which cover out-of-pocket expenses for moderate-income households. It reports that these subsidies will "cost about $7 billion this year and $10 billion in 2018."

It would have been useful to put these numbers in a context that would be meaningful to readers. The 2017 figure is equal to roughly 0.18 percent of total spending and the 2018 number would be roughly 0.25 percent. On a per person basis, the 2017 number is equal to about $21 for every person in the country, while the 2018 number would be around $30 for every person in the country.

Add a comment

The NYT had an interesting piece on how Japanese workers are increasingly working well into their sixties, as a declining population has led to somewhat of a labor shortage. The piece rather bizarrely offers this as an explanation for why wages aren't rising, since it says that older workers are paid less.

That could be true, but it would imply serious discrimination if older workers are paid less than younger workers with the same productivity. Alternatively, if older workers are paid less because they are actually less productive, then this would not explain why average wages are not rising in the face of a labor shortage.

It is also worth noting that the piece repeatedly describes the declining numbers of workers as a problem. This is 180 degrees at odds with the view that robots are going to take all the jobs and we won't have any work for people. It is incredible that we have ostensibly serious people who are both worried that an aging population will leave us with too few workers and rapid productivity growth (i.e. robots) will leave us with too few jobs. As they say in economics, "which way is up?"

Add a comment

The NYT had an interesting column making the case for publicly funded open research to speed the development of artificial intelligence. It's good to see some clear thinking about alternatives to research supported by government-granted patent monopolies. Can we talk about prescription drugs now?

Add a comment

If the United States is to have more rapid economic growth, as most folks seem to want, then it needs more rapid productivity growth. Productivity growth is the key factor allowing rising living standards through time.

This is basically definitional. It means more output of goods and services per hour of work. It can allow us to have more stuff or work fewer hours and have the same stuff.

When we consider this simple logical point, it is bizarre that one of the attacks against Amazon is that it is costing jobs in retail. If Amazon is costing jobs in retail, it is increasing productivity in the sector. It is allowing us to get by fewer workers, meaning that workers can instead be available to do things like provide child care, install solar panels, or do many other useful things. Or, we can all work less.

Of course, they may not be the same workers and some people may get screwed in the process, but this is a larger problem of policy. For example, the Federal Reserve Board is quite deliberately trying to slow the rate of job creation by raising interest rates. Let me repeat that since it seems many people involved in policy debates somehow haven't noticed: the Federal Reserve Board is quite deliberately trying to slow the rate of job creation by raising interest rates.

If Amazon and other online sellers were not eliminating jobs in retail, the Fed would be raising interest rates faster in response to a more rapid rate of job creation. Insofar as increased productivity in retail is slowing the rate of job creation, it is allowing for jobs to be created in other sectors which would not otherwise exist.

It is also worth noting that productivity growth has been extraordinarily slow in the last decade, as in the opposite of fast. This is the main reason for the slow growth in the recovery. In other words, the economy's main problem has been too few job-killing robots.

(Yes, distribution is a problem and we should talk about things like patents and protectionism for doctors and other professionals, but let's keep it simple for now.)

Add a comment

That's the arithmetic according to the Washington Post. The article told readers that Wisconsin governor Scott Walker claims he is giving $3 billion in economic incentives to the Taiwanese manufacturer Foxconn. The company, in turn, says that it will build a factory that will eventually employ 3,000 people. It would have been helpful to include the arithmetic so it was clear how much Scott Walker planned to spend for each job that Foxconn claims it will create. (Foxconn has not always followed through on its promises, as the piece points out.)

It also would have been worth mentioning the importance of these jobs to Wisconsin's labor market. The state currently has 472,000 manufacturing jobs (it added 7,900 in the last year), so the jobs pledged by Foxconn would increase this figure by just under 0.6 percent.



As several comments noted, the Washington Post had a follow up piece that gives more details on this deal. The 3,000 jobs figure is the number who are originally expected to be employed in the factory when it is projected to be completed in 2020. Foxconn said the number of people employed has the "potential to grow" to 13,000. It appears that the full $3 billion in subsidies is at least partly contingent on the number of jobs actually created, although the terms do not seem to be clear at this point.

Add a comment

Morning Edition host Rachel Martin introduced an interview with Tom Nickels, spokesperson for the American Hospital Association by referring to Republican efforts to "fix" the nation's health care system. This implies that the Republicans are trying to make the system better. This is certainly not obvious since all of their proposals are likely to take away insurance from tens of millions of people.

It shouldn't be too hard to use neutral terms, like "change" or simply refer to efforts to repeal the Affordable Care Act. There is no reason for NPR to attribute good motives to the Republicans, especially when there is zero evidence to support this view of their actions.

Add a comment

At a speech in Youngstown Ohio last night, Donald Trump talked about the loss of manufacturing jobs in the state and told his audience:

"They’re all coming back. They’re all coming back. They’re coming back. Don’t move. Don’t sell your house."

Actually, they were coming back (at least some of them) before President Trump took office, but the state is again losing manufacturing jobs.

Manufacturing Employment in Ohio

OH Man

Source: Bureau of Labor Statistics.

Employment in manufacturing in Ohio had increased by 4,700 in the year from January 2016 to January 2017. In the five months since January 2017, it has fallen by 5,400. If this rate of job loss continues, Ohio will lose almost 70,000 jobs manufacturing jobs, more than 10 percent of employment in the sector, over a two-term Trump administration.

Add a comment

Neil Irwin had an interesting Upshot piece highlighting a new paper by J.W. Mason arguing that slow productivity growth is in large part due to slow GDP growth. The basic argument is that if growth were faster, labor markets would be tighter, and companies would have more reason to invest in labor saving equipment.

While this argument strikes me as undoubtedly true, there is another aspect to productivity growth that is often missed. One thing that is even easier than replacing workers with equipment is simply not replacing workers. In other words, most employers can run stores, restaurants, or other businesses with fewer workers. The cost of this is likely to mean that customers have to wait longer to be served.

This could mean, for example, that when you get to the checkout counter at a supermarket you have to wait ten or fifteen minutes in line rather than having someone immediately available immediately to serve you. The same would apply to lines at fast food restaurants or the pace of service at a sit-down restaurant. Instead of having workers available for customers at all times (which means they do nothing, some of the time), employers will make customers wait.

This would show up as an increase in productivity as conventionally measured. Output would be unchanged, but fewer workers are employed than in the good service scenario. In principle, if we have perfect productivity data, this would not be the case, since the longer wait times should be reported as a deterioration in quality and therefore a price increase, which would mean lower output. But we don't have perfect data, so in our productivity numbers, longer wait times mean higher productivity (and vice versa). 

Add a comment

Binyamin Appelbaum had an interesting discussion of inflation in the NYT yesterday. As he notes, it has been below the Fed's target throughout the recovery and, contrary to expectations, it has been falling in recent months. This suggests that the economy could be operating at a higher level of output with more employment. That would put more upward pressure on wages and lead to somewhat higher inflation. That suggests that the Fed may have been wrong in its recent interest rates hikes which were intended to slow growth.

There are a few other points worth noting about inflation while we are on the topic. While it is common to say that inflation hurts investment, at least in the U.S. this does not appear to have been the case.



As can be seen the investment share of GDP peaked in the late 1970s and early 1980s when inflation was also running at its most rapid pace in the post-World War II era. Investment has been considerably lower as a share of GDP in the last three decades of moderate inflation. The one exception when investment got close to its peak of the high inflation era was at the end of the 1990s during the stock bubble.

Add a comment

You know the person who commits murder and the dead person really are both victims in Robert Samuelson land. His latest column on health care shows his great expertise in obscuring everything he touches to say it's all just so complicated.

He tells readers:

"Still, there’s no moral high ground. Some Democrats have wrongly accused Obamacare opponents of murder. This is over-the-top rhetoric that discourages honest debate. It’s also inconsistent with research. Kaiser reviewed 108 studies of the ACA’s impact and found that, though beneficiaries used more health care, the 'effects on health outcomes' are unclear."

Yes, Kaiser was being very careful in its comments. The Affordable Care Act has been in effect for three and a half years. There is a lag between research and publication, that means that at this point in time we still have limited solid measurements of outcome measures. We do have data on diagnoses and treatment.

The study reports:

So we have evidence now that people with conditions like heart disease and cancer are being diagnosed earlier and getting treatment. We are not going to have good data on mortality rates at this point since the vast majority of people with heart disease and cancer do not die immediately from these conditions. But, if we make the huge leap that treatment might affect survival, then we can infer that the ACA is keeping people from dying.

But hey, we want to be cautious, unlike those irresponsible Democrats who are accusing the Republicans of murder. After all, it is possible that all the money we spend on treating heart disease and cancer is totally worthless.

Add a comment

The Washington Post has devoted enormous resources to trying to convince its readers that the federal government's disability programs are in crisis. And it has no qualms about misrepresenting the data to make its case.

Today we got a great example in a question and answer session in reference to its latest major feature piece. In answer to the question, "what's the problem?" it tells readers:

"The program for disabled workers, which Congress had to rescue from insolvency in 2015, is estimated to go broke again sometime over the next decade or so. The government this year is expected to spend $192 billion on disability payments — more than the combined total that will be spent on welfare, unemployment benefits, housing subsidies and food stamps."

The assertion that the program will go broke is extremely misleading. Even if Congress never did anything it could still pay will over 90 percent of projected benefits for more than two decades into the future and even at the end of the 75-year planning period, it is still projected to be able to pay over 80 percent of scheduled benefits.

This is an important point since many politicians have advocated cutting benefits to keep the program fully funded. If the point is to ensure to prevent benefits from being cut due to a shortfall, cutting benefits to make up the gap doesn't help.

It is also worth noting that the $192 billion figure includes the Supplemental Security Income (SSI) program which is funded out of general revenue, not a payroll tax. While it can make sense to combine the programs as disability programs, in doing so it would be worth noting the third program in this category, workers' compensation. Most states have substantially reduced the generosity of their Workers Compensation programs over the last three decades. As a result, the total amount spent on disability as a share of GDP has not increased by very much over this period.

The comparison to "welfare, unemployment benefits, housing subsidies and food stamps" is also misleading, since we actually spend very little on these programs even though the public perception is that they comprise a large share of the budget. The Post presumably knows the public hugely overestimates the share of the budget spent on these programs so why would it use them as a base of comparison, unless the point is to create the impression that these disability programs are a very big share of the budget.

It actually would not be hard to convey the spending in a way that would be meaningful to most readers. It is equal to roughly 1.0 percent of GDP. It is a bit less than 5.0 percent of total federal spending. Alternatively, it is a bit more than 30 percent of projected military spending for 2017.

Add a comment