Blog postings by CEPR staff and updates on the latest briefings and activities at the Center for Economic and Policy Research.

Back in the 1990s stock bubble it was common for analysts to say things like price-to-earnings ratios (PE) no longer mattered. They were right, at least for a while, as the stock valuations of companies like AOL and Priceline soared way beyond anything that could conceivably be justified by current or future earnings.

Of course after a while, price-to-earnings did come to matter, as the stock market lost half its value from its peak in March of 2000 to its trough in the summer of 2002. The tech heavy Nasdaq lost close to 80 percent of its value. Many of the big tech enthusiasts were wiped out in this crash. While it might seem old-fashioned, people presumably value stock based on how much earnings a share commands, not the beauty of the stock certificate or how cool the company is.

With this in mind, it is interesting to think about what the Amazon future might look like given that it now has a market capitalization of roughly $480 billion with current profits of roughly $2.6 billion. This gives it a price-to-earnings ratio of 184 to 1.

Add a comment

While business cycles vary in their length and growth rates, there are some consistent patterns. Most obviously, the labor market tightens as the business cycle advances with unemployment falling and the percentage employed rising. This tightening of the labor market increases the bargaining power of workers since they are more likely to have a choice of jobs than they did during the downturn or during the early phase of the recovery. As a result, workers are likely to move to more desirable and better-paying jobs. Better paying jobs are also likely to be more productive jobs, which mean that the shift in employment patterns as a result of a tightening labor market could provide some boost to productivity growth. (This is offset in part by the fact that large numbers of people shifting jobs will reduce productivity.)

This pattern is likely to be especially strong among less-educated workers since they are the ones most likely to lose their jobs in the downturn. An employer is far more likely to lay off a retail clerk or assembly line worker than a store manager or a shift supervisor. This means that the improved labor market situation during the upturn is likely to disproportionately benefit workers at the bottom end of the wage distribution.

Add a comment

As discussed in a recent Federal Reserve staff working paper, “recessions may impact different groups at different phases of the aggregate business cycle”. The paper finds that in an economic downturn jobs losses disproportionately hit black and Hispanic workers relative to white workers, while periods of unemployment likewise last longer for black and Hispanic Americans. As a result, in the later stages of an economic recovery new jobs are added at a more rapid rate for black and Hispanic workers.

An important monetary policy implication of this finding is that the gap in labor market outcomes for different groups is affected by how quickly the Fed raises interest rates in an economic recovery. Put another way, if the Fed moves too quickly to raise interest rates, it will disproportionately leave black and Hispanic Americans out of the labor market, and particularly black and Hispanic women and youth.

Add a comment

The unemployment rate rose very slightly to 4.4 percent in June, as the estimated size of the labor force expanded at its fastest rate (361,000) since last July (406,000). This is the third consecutive June in which the labor force estimate reversed a big change in May. Thus, despite the rise in the unemployment rate, the 245,000 net new jobs in June raised the overall employment-to-population ratio from 60.0 percent in May to 60.1 percent in June.

The establishment survey showed further evidence that last month’s seeming weakness was a little misleading. Job growth rose to 222,000 in June, and these nonfarm payroll jobs for April and May were revised up by 47,000.

Finally, average hourly wages ticked up by 4 cents in June, and 2.5 percent over the last twelve months.

Thus, people continue to return to the labor force as jobs continue to be made available. This suggests that despite low unemployment there is room for growth.

Add a comment

President Trump and Republicans in Congress have repeatedly charged that the Affordable Care Act (ACA) is collapsing. They point to insurers dropping out of the exchanges and endlessly cite the fact that more than 1300 counties across the United States only have one insurer operating in the exchanges and that some will not have any in 2018.

The lack of competition in the exchanges is a serious problem. While people can still buy insurance in the individual market off of the exchange, and still benefit from the ACA prohibition against discrimination based on pre-existing conditions, they are not eligible for ACA subsidies unless they buy insurance through the exchanges. These subsidies are necessary to make insurance affordable for millions of people.

So, the lack of a vibrant market in many counties is a serious issue for the ACA. However, there is an important part of the story that Trump and other Republicans forget to mention. The lack of competition in the exchanges is overwhelmingly a problem for people living in states controlled by Republican governors.

The graph below shows the number of people living in counties that only have one insurer in their exchange by the party of the state’s governor.


As can be seen over 40 million of the people in counties with only one insurer in the exchanges live in states with Republican governors. By comparison, just 10.7 million people who only have one insurer in the exchanges live in states with Democratic governors.

Add a comment

Prescription drugs are a large and growing share of national income. While it is generally recognized that drugs are expensive, many people are unaware of how large a share of their income goes to paying for drugs because much of it goes through third party payers, specifically insurance companies and the government.

The Centers for Medicare & Medicaid Services (CMS) produce projections of national expenditures on prescription drugs through 2025, along with historical estimates dating back to 1960. As shown below, prescription drug spending from 1960 to 1980 was equivalent to about one percent of total wage and salary income. In the years leading up to the passage of the Bayh-Dole act in 1980, wage income was rising faster than spending on prescription drugs. As a result, the share of wages spent on prescription drugs was actually falling, reaching a low in 1979 of 0.86%.

Prescription Drug Spending Relative to Wages
Add a comment

Which industry you work in has a large effect on the extent of living standard improvement you've seen during the 21st century, according to data on the earnings of production and non-supervisory employees. Since 2000, the growth in workers' purchasing power (their ability to use earnings to buy goods and services) has varied strongly by industry sector. Per hour worked, financial sector employees are able to purchase 24.6% more goods and services than they could in 2000, while manufacturing employees have attained only 2.5% percent total growth in purchasing power over the 17-year period. Workers in the retail trade (-0.3%) and transportation and warehousing (-0.6%) sectors have seen their purchasing power deteriorate.

Worker Purchasing Power Growth by Industry

Much of the difference depends on the overall health of the economy as workers in some industries are more vulnerable to swings in the economy. During an economic downturn or periods of low economic growth, manufacturing workers are more likely to be laid off than doctors. This tendency has direct consequences for the manufacturing workers who are not laid off, as the availability of unemployed workers makes it possible for employers to offer fewer or smaller wage increases without their employees quitting. In cases where workers can easily move across industries, job loss in one industry will lead to lower wages in related industries.

So how is it possible that the workers in the industry sector that gave its name to the recent global crisis, financial activities, have the largest purchasing power increases? One possible explanation comes from the presence of strong government support for the financial sector during the economic downturn. Unlike the non-interventionist approach towards protecting vulnerable workers from falling wages, taxpayer resources were used to spare the financial sector from the contractionary portion of the economic cycle.

Add a comment

Many economists have expressed surprise over the fact that investment has not been stronger in the recovery from the Great Recession, given the high level of corporate profits. The trends in investment and profits over the last half century suggest that they should not be surprised.

Profit and investment shares of GDP have not moved together over this period. In fact, there is weak negative correlation in shares over this period (-0.3). The figure below shows before- and after-tax shares of profit in net national product (NNP) since 1964.[1] (We use NNP to take account of the fact that the depreciation share of output has increased substantially over this period, which would bias the profit share downward.) It also shows the share of non-residential fixed investment in NNP.

There are several points worth noting about these trends. First, there is not much movement in the non-residential investment share. The table below shows the investment shares of NNP by decade:

Decade Investment Shares
1970s 13.9%
1980s 15.8%
1990s 14.6%
2000s  14.9%
2010-2017(1)  14.3%
Add a comment

The unemployment rate fell to 4.3 percent in May, a new low for the recovery and the lowest level since 2001. However, this decline in unemployment was the result of people leaving the labor market; as the number reported as employed in the household survey actually fell, with the overall employment-to-population ratio dropping from 60.2 percent in April to 60.0 percent in May.

The establishment survey showed further evidence of a weakening labor market as the pace of job growth slowed to 138,000 in May. There were also substantial downward revisions to the prior two months’ job growth numbers, which brought the average for the last three months to just 121,000.

In addition, wage growth appears to be moderating rather than accelerating. The year-over-year increase in the average hourly wage is just 2.4 percent. Taking the average of the last three months compared with the average of the prior three months, wages are rising at just a 2.2 percent annual rate.

There is certainly little evidence in this report that the labor market is overheating or is likely to do so any time in the foreseeable future.

Add a comment

While rising capital share and greater concentration of wealth explain some of the story of economic inequality, the largest part of the story is the growth in wage inequality over the last several decades. Available data from the Social Security Administration unfortunately doesn’t go past 1990, overlooking considerable upward distribution of wages beginning in 1980. However, wage distributions from 1990 to 2015 show a clear, and unequal, upward trend.

The share of wages earned by the top 0.1 percent of wage earners increased 36 percent in that time period, from 3.5 percent of all wages earned to 4.8 percent. These earners are largely Wall Street bankers and top executives from private companies, as well as hospitals, universities, and other non-profits. Although the data from such a small pool of workers is erratic, they show soaring gains over ordinary workers that coincide with stock market peaks. Wages at this income level are likely paid in part in stock options, so that connection is unsurprising, but the magnitude of wage increases for this group compared to the others supports the argument that wages are part of the inequality picture.

Add a comment

On the 17th of this month, a group of House Democrats, including Representative John Delaney (D-MD.), delivered a letter to President Trump offering, essentially, a trade: A tax holiday for international corporations in exchange for the guarantee that the money from that repatriation would be used exclusively to fund the country’s long-overdue infrastructure maintenance projects. Earlier this year Rep. Delaney also authored a tax and infrastructure bill which would allow corporations with funds outside the U.S. to return that money to the country at a tax rate of 8.57 percent (instead of the top corporate tax rate of 35 percent). Delaney, in his statement, called the bill a “pro-growth reform” and his co-sponsor, Rep. Rodney Davis (R-IL), said that it would “spur job creation”.

Unfortunately, the Senate Subcommittee on Investigations has looked into this proposition and it disagrees. In 2011 the committee analyzed the Bush tax holiday (which also promised to increase job growth and boost the economy) and found that it did the exact opposite. Not only did the tax cuts lead to a net decrease in hiring by the companies that took advantage of them (which were almost exclusively in the pharmaceutical and tech industries), the money went predominantly to stocks and executive bonuses rather than new investments. The cuts also cost the Treasury an estimated $3.3 billion in revenue over the subsequent ten years. There’s no reason to think the results of a tax holiday would be any different now, and that lost tax revenue only exacerbates long-term federal funding shortages for infrastructure and other critical projects.

Add a comment

In March, the Federal Reserve Board voted to raise interest rates for the second time this year, indicating their concern that rising wages would result in inflation if they did not take action. However, an analysis of average hourly wage trends for six sectors of employment since 2010 finds little evidence that wages are rising dangerously fast, or even accelerating at all.

Add a comment

The overall unemployment rate fell to 4.4 percent in April, tying the lowest level reached since May of 2001 as the establishment survey reported the economy added 211,000 jobs. With roughly offsetting revisions to the prior two months' job growth, this brings the average for the last three months to 174,000.

While the report on the whole is quite positive, one item is especially worth noting – the increase in the employment-to-population ratio (EPOP) among prime-age men. The employment rate for prime-age workers edged up to 78.6 percent. This is a new high for the recovery, although it is 1.7 percentage points below the pre-recession peak and 3.3 percentage points below the 2000 peak.

In April the EPOP for prime-age women edged down to 71.9 percent, 0.9 percentage points below its pre-recession peak and 3.0 percentage points below its 2000 peak. However the EPOP for prime-age men rose 0.2 percentage points to 85.4 percent. This is still 2.6 percentage points below its pre-recession peak and 4.1 percentage points below its 2000 peak, but the rise does suggest that there is still more room for EPOPs to increase among prime-age men. The fact that the EPOP for both prime-age men and women remain below pre-recession levels and far below 2000 levels strongly suggest that the issue is a lack of demand in the economy and not a decrease in the ability or desire of people to work.

Add a comment

Earlier this month the Washington Post’s editorial board published a criticism of the Social Security Disability program. Noting the decline in labor force participation amongst prime-age men, the board suggests that the disability program discourages its recipients from working. The article argues that the sudden loss of cash benefits (and affordable health insurance) when recipients rejoin the workforce might persuade many to remain on disability when they are able to work.

It seems unfair to lump these two factors together, since the program’s modest monthly benefits are far less than the wages of any near full-time job, while the loss of health insurance, especially for disabled workers who may require additional services, would be devastating. The Post is probably right to say that health insurance costs play a strong role in workers’ choices about their employment level, but that suggests a different type of reform.

Add a comment

Dean Baker, Rebecca Vallas, Katherine Gallagher Robbins, and Rachel West have already pointed out most of the problems with Washington Post’s recent story and its subsequent editorial on Social Security Disability Insurance (SSDI). But I want to say a bit more about one aspect of how the Post’s framing of the story, and use of cherry-picked data, paints a misleading picture for the public.

The Post tells the story of a 39-year-old man living in a small Alabama town who is deciding whether to apply for SSDI. The accompanying data hypes two things: the increase in the number of people ages 18-64 receiving disability benefits since 1996 and the disability beneficiary rates by county since 2004. (As West and Gallagher Robbins found, the Post made several errors in their analysis, some of which they corrected, but the data errors aren’t my focus here). The editorial cites the initial story, links SSDI to the declining labor force participation (LFP) of prime-age men, and calls for cutting SSDI to “make sure work pays for all who are willing to work.”

The story and the editorial are only the most recent of elite national media pieces promoting the idea that SSDI is a de facto unemployment assistance program for prime-age men who may be “desperate” but aren’t really disabled. Reading these stories, I imagine many readers come away with the impression there has been a large increase in the number of 30- and 40-something men who are receiving disability benefits, and that this is driving a large part of the decline in prime-age men’s labor force participation over the last two decades.

Add a comment

President Trump has promised to release his tax plan this Wednesday, April 26. While details are lacking, one thing we can be sure of is that it will include a reduction in the tax rate on so-called ‘pass-through’ business income. Pass-through income is the net income of a business that passes through to a business’ owner or partner and is taxed at that individual’s tax rate. The income received by an individual who is the sole owner of a business like a bakery or deli after all expenses are paid is an example of pass-through income. So it is the net income of high flying partnerships that is distributed to the wealthy partners in private equity firms, hedge funds, and Wall Street law or tax accounting practices. President Trump, who is a partner in hundreds of real estate ventures and other businesses, receives pass-through income.

Add a comment

Last month the Congressional Budget Office (CBO) announced that the (now defunct) American Health Care Act (AHCA) would have reduced off-budget spending by $5.9 billion in 2026. Off-budget spending, otherwise known as Social Security, isn’t mentioned in the bill once, so how could the AHCA affect its spending? The answer is that the bill, in CBO’s assessment, would have discouraged older workers from retiring or moving to part-time work and starting to collect Social Security benefits.

Under the Affordable Care Act, workers have the flexibility to obtain health insurance from the exchanges instead of relying on employer-provided plans. Without it, we can infer that CBO’s projected savings came as the result of 500,000[1] older workers delaying their retirement, and therefore not collecting Social Security benefits in 2026.

Add a comment

The unemployment rate fell to 4.5 percent in March, its lowest level since May of 2007. The employment-to-population ratio also edged up to 60.1 percent, a new high for the recovery, but still more than 3.0 percentage points below its pre-recession level.

However, the good news on the household survey was accompanied by weak job growth in the establishment survey. The economy added just 98,000 jobs in March. Job growth was also revised downward by 38,000 for the prior two months, bringing the three month average to 178,000. There also has been some shortening of the average workweek. The index of aggregate weekly hours is unchanged from its January level.

The strongest areas of job growth were restaurants (21,700), building support services (16,800), and health care (13,500). Mining also added 11,000 jobs, as did manufacturing. Retail was a big job loser in the month, shedding 29,700 jobs. This sector is likely to continue to show weakness as several major chains have announced plans to close a large number of stores.

Wage growth appears to be slowing slightly. While the year-over-year increase in the average hourly wage was 2.7 percent, wages have grown at just 2.4 percent comparing the average of the last three months to the prior three months. This should give pause to those concerned about the labor market being too strong. The fall in the length of the workweek, coupled with modest wage growth, indicates there is much room for further strengthening.

Add a comment

Data from the Consumer Expenditure Survey show that the bottom 60 percent of the US population has been spending over 30 percent of their income on housing since before 1989. In 1981, the Department of Housing and Urban Development (HUD) issued a recommendation that a maximum of 30 percent of household spending go to housing costs. This clearly has not been the norm for most households over the last three decades.

For the lowest income quintile (with an average income of $10,916 in 2015), rising housing costs are an especially large burden, averaging over 40 percent of income spent on housing each year since 2007. At the other end of the income spectrum, the top income group (average income of $177,851 in 2015) spent more than recommended on housing for a majority of the years analyzed, although the most recent numbers are back below 30 percent.

Add a comment

Unemployment rates for black workers continue to be well above those of white workers – within any given age group, the black unemployment rate is often double the white unemployment rate. In addition to being much higher, black workers’ experiences with unemployment are typically different from white workers’ unemployment experiences.

This post analyzes unemployment data from 2016 to gain some insights into the nuances of black unemployment – underemployment, length of unemployment, and causes of unemployment. For context from previous years, see this 2015 CEPR post on racial inequalities in unemployment.

Add a comment

Tuesday night, Rachel Maddow and David Cay Johnston revealed Donald Trump's 2005 two-page Form 1040 on air. Before revealing the form, Maddow made an extended argument that Donald Trump must release his complete tax returns in order to disclose all potential conflicts of interest from his business empire, including potentially income from foreign governments. 

We agree — but tax returns are merely a starting point for understanding Trump’s business partners.

In one example, Maddow pointed to a sketchy real estate deal with a Russian oligarch. Donald Trump purchased a piece of property in Florida for $40 million in 2005. Just three years later, Trump sold the property for $100 million to a Russian oligarch (i.e., a rich Russian businessperson closely tied to Putin’s government) named Dmitry Rybolovlev.

Trump's 150% return on investment in just three years would be suspiciously large even if the real estate magnate had bought an undervalued property and improved it. However, Rybolovlev actually quickly tore down the 62,000 square foot mansion and sold it off in three pieces. How the value of the property appreciated so quickly is a mystery, especially as the Florida’s real estate market was collapsing.

Add a comment