Dispelling the Myth of Low Unemployment

Doing some weekend catch-up, and a reader pointed me to a very good post by Barry Ritholtz, which confirms something I’ve believed but haven’t gone to the trouble to prove, namely, that unemployment is much higher than the government releases would have one believe.

Full disclosure: I’m skeptical of quite a few government stats: GDP (badly tainted by hedonic adjustments), inflation (calculation methods revised in the 1990s which reduced annual reported CPI roughly 0.5%, which lowers Social Security and other CPI benefit adjustments; Ritholtz has been active on this beat), and more recently, job creation figures.

As Ritholtz says,

Anyone with more than 4 functioning brain cells should be able to figure out that a 4.5% unemployment rate would be causing huge labor shortages and wage increases.

On a macro level, if unemployment were as low as 4.5%, labor would have more bargaining power, which is inconsistent with widely reported stagnant wages.

On a micro level (yes, this is anecdotal), there are way way too many people over 40 who are un- or underemployed, or “retired” at a modest level when they’d rather be working. If you lose that corporate meal ticket, it is an long way down. In most cases, mid to senior corporate employees and professionals have narrow skills that can only be deployed in roles similar to the one they lost. And companies are onservative in how they hire (another sign of a slack labor market, since if it were hard to find workers, they’d have to accept employees who were a less than exact fit).

One of many tidbits from the other end of the employment spectrum: over 5000 people applied for 300 jobs at Apple’s new Fifth Avenue store in Manhattan. Now perhaps many of them were employed in retail elsewhere and wanted what they thought was a kinder and cooler environment, but no matter how you dress it up, this is a on-your-feet-all-day, low-end job. Note that other large store openings in the New York metro area get similar overwhelming responses even though the BLS says the unemployment rate here is on par with the national level.

Back to Ritholtz:

So long as we are popping economic myths, let’s also dispatch with the 4.5% unemployment rate. That number has been largely caused by several million exhaustees and others simply leaving the work force:

The actual unemployment rate is closer to 6.5%. And if we measured it the way the Europeans do, its closer to 8%. This explains why wages and labor costs have remained subdued despite the alleged 4.5% UE measure.

Print Friendly, PDF & Email


  1. Aron Roberts

    Yves, thank you for your many trenchant posts. Your blog is a recent discovery – and a treasured one, along with Calculated Risk, and several others blogs covering similar topics – in helping make sense of the recent perturbations in financial markets.

    Regarding US consumer price inflation, as measured by the Bureau of Labor Statistics’ CPI variants, you might find of interest John Williams’ “shadow statistics”, http://www.shadowstats.com/.

    Williams believes that, if the BLS calculated inflation using the same manner that they did back in 1980, the last year of the Carter Administration, we’d have been seeing 8-10% year-over-year official CPI over the last decade (see http://www.shadowstats.com/imgs/sgs-cpi.gif); if it were calculated the same as it was in 1990, annual increases would have been in the 4-7% range (see http://www.shadowstats.com/imgs/chartSGS01-35.gif).

    Williams’ “reworkings” of the CPI are of course subject to dispute, but there is ample evidence that multiple statistical adjustments since 1980 have cumulatively reduced reported CPI, making apples-to-apples comparisons with past, official inflation rates extremely difficult.

  2. Steve

    I’d recommend taking anything Barry Ritholtz says with a boulder of salt.

    For example, this quote:
    “The actual unemployment rate is closer to 6.5%. And if we measured it the way the Europeans do, its closer to 8%. This explains why wages and labor costs have remained subdued despite the alleged 4.5% UE measure.”

    No sources, no evidence to back it up – it’s just there. And yet people gobble it up. (I especially like his nebulous reference to “the Europeans.”)

    The OECD has a data series call the Standardized Unemployment Rate (SUR) that puts the unemployment rates of countries on equal footing. You can see it here: Link

    As you may guess by now, it confirms that Barry Ritholtz is full of it.

  3. Yves Smith

    Ritholtz did provide a chart which shows declining labor force participation. And if you are familiar with how the BLS conducts its surveys, “marginally attached” workers, which include “discouraged workers, are NOT counted as unemployed. They are classified as not in the labor force at all.

    I have not been able to check extensively, but the only paper I found so far that addresses that issue indicates that in Italy, 90% of discouraged workers are classified as unemployed (t only mentioned the US, Australian, and Italian definitions. Australia’s definition of a discouraged worker is a bit more restrictive that ours and would lead to higher unemployment figures). That would seem to support Ritholtz’s view.

    If this bothers you, I suggest you take it up with him.

  4. Steve

    From the BLS link you posted:

    Who is counted as unemployed?

    Persons are classified as unemployed if they do not have a job, have actively looked for work in the prior 4 weeks, and are currently available for work.

    And now from the European Labor Force Survey found here:

    Unemployed persons are persons aged 15-74 (in ES, NO[3], SE (1995-2000), UK and IS 16-74) who were without work during the reference week, were currently available for work and were either actively seeking work in the past four weeks or had already found a job to start within the next three months.

    Sounds pretty similar. And since discouraged workers are NOT looking for work, they would not be considered unemployed in the European Labor Force Survey.

    If this bothers you, I suggest you take it up with him.

    Sorry, I assumed this topic was open for discussion here since an article was posted about it.

  5. Steve

    One last thing, the BLS DOES track alternate measures of unemployment. And one of those includes discouraged workers. It’s called the U-4 unemployment rate and you can see it here.

    So even if you believe that discouraged workers should be counted as unemployed, counting them only leaves you with a 4.9% unemployment rate, not the 6.5% Ritholtz suggets. If you want to go ahead and include marginally attached workers, look at the U-5 number. That’s still only a 5.5% unemployment rate.

    So here is something that can be easily looked up with a simple trip to the BLS website, yet he makes up a number. Why? I have no idea…

  6. Anonymous

    The Bushies will always have their defenders. That’s what brainwashing does to you.

    It is significant that even Reagan Republicans like Paul Craig Roberts have turned against Bush because they figured out early on what a poseur he and gang of Republicans are today. He has a collection of articles on Counterpunch.org with numbers from the BLS where he dissects them.

    Another interesting thing about the Bushies has been without fail that once quarterly economic & labor numbers are released (in which they claim are “spectacular” by their measure, but underwhelming by any previous standard, even Jimmy Carter’s), then 2-3 weeks they’re always (without fail) revised downward (in some cases by half).

    Back in 2000, I was calling Bush George W Hoover. Why? Because despite all of Rove’s claims, even the least bit of homework would have turned up the fact that his state-level economic plan was hogwash. Stuff like funding public schools with taxes on strip clubs.

    I think after Bush, this makes my case for giving Texas back to Mexico stronger. Afterall, it’s produced LBJ, GHWB and GWB. Of course we’ll have to tell Mexico we were wrong to take it, but that’s minor compared to the damage people from this state have done to the country.

    Wanna hear the big myth lanced?

    There is a widely held belief that Republicans are better for business than are Democrats. Let’s look at the facts. The wild stock market ride of recent weeks does not compare to the two worst stock events, the crash of 1929 and the 1987 free fall, which also occurred under Republican administrations. Since 1900, Democratic presidents have produced a 12.3% annual return on the S&P 500, Republicans only 8%. Gross Domestic Product growth since 1930 is 5.4% for Democratic presidents and 1.6% for Republican presidents.

    Bush inherited from President Clinton an annual federal budget surplus of $236 billion, the largest in American history. Clinton balanced the budget for the first time since 1969. Budget surpluses were expected to total $5.6 trillion between fiscal year 2002 and 2011.

    Despite this, Bush transformed the surpluses into a $1.1 trillion annual deficit in just three years because of the Iraq war and his relentless push for permanent tax cuts for wealthy Americans, a new iteration of Herbert Hoover’s equally catastrophic “trickle-down” theory. Bragging about a $239 billion deficit sets such a low standard that Bush can claim horrific failure as a good thing for the country. The Bush administration’s annual loss of three-quarters of a trillion dollars is unprecedented. Bush presided over the loss of 2 million American jobs in his first 2 1/2 years and has net gained 5.6 million in six years, the worst since Hoover. Clinton created 23 million jobs.

    Economic Indicators: Democratic Versus Republican Presidents

    In six major criteria – GDP growth, per capita income growth, job creation, unemployment reduction, inflation reduction, and federal deficit reduction – for the ten post-World War II presidencies until Bush, there is a record to track the reality of Democratic versus Republican economic success.


    * Lyndon B. Johnson’s “Great Society” created robust economic expansion, first in both GDP and personal income growth. He also reduced unemployment from 5.3% to 3.4%. Economic growth remained robust through most of LBJ’s presidency.

    * John F. Kennedy campaigned on the idea of getting America moving again, and he did. Under Kennedy, America entered its largest sustained expansion since WWII. GDP and personal income growth were second only to Johnson, all with minimal inflation. Contrary to Republican attempts to say Kennedy’s tax cuts are like Bush’s, Kennedy’s were targeted at middle and lower incomes.

    * The economy added 10 million jobs under Jimmy Carter despite high inflation; Carter ranks first in job creation next to Clinton during just four years in office. Carter also reduced government spending as a percentage of GDP.

    * Harry Truman’s second term saw the fastest GDP growth and the sharpest reduction in unemployment of any president surveyed (of course, FDR’s post Hoover-depression New Deal jobs are first).


    * Ronald Reagan focused on reducing the cost of capital through cutting tax bracket highs for the rich and reducing the size and scope of government. But, instead of lowering spending, Reagan shifted money to the military (i.e. Star Wars) and the deficit tripled with the tax cuts and military spending – as under Bush II.

    * Under Gerald Ford, the deficit soared and the unemployment rate grew from 5.3 – 8.3% in just 2 years. His “WIN” (Whip Inflation Now) buttons were no match for economic inactivity.

    * It was under Richard Nixon that inflation started to spiral out of control, from 4.4% to 8.6%, and the deficit shot up from $2.8 billion to $73.7 billion.

    * The Eisenhower years were characterized by slow growth (2.27% annualized GDP growth) and relatively high unemployment (7.7% at end of term).

    * George H. W. Bush had the poorest record for both GDP and income growth. During his single term, the deficit ballooned (from $152 billion to $255 billion) more than under every president but his son and Ford.

    (Sources: White House Office of Management and Budget (OMB), U.S. Department of Labor (DOL), and White House Council of Economic Advisors)

  7. Anonymous

    I also find it amusing that in three places (Barry Ridholtz’s article, Yves article, and my previous comment in a prior post) we’ve all three said in effect the government numbers have been manipulated (Shadowstats also says that too in the archived articles).

    So of course the Bushies point to the government numbers their proof.

    I suppose we’ll have to listen to readings of “My Pet Goat” as proof that they are indeed literate, but they still won’t comprehend the meaning, as with all other things. ;)

  8. Yves Smith


    Perhaps the reason that Ritholtz doesn’t cite other data is that the BLS is probably the only source available for the numerator and denominator of the unemployment figures. However for other government statistics where there has been a gap between observable reality and the government releases, Ritholtz and others have correctly pointed to disparities, and have been able to point to corroborating information. For example, Ritholtz and others have repeatedly, and correctly, called it when the initial GDP releases were off (and I don’t mean by a little). He and others have written about the dubioius “birth/death” model in the monthly job creation releases, which for the last few months have shown significant increases in employment in construction in the face of a tanking housing market. He has taken a view on “inflation ex inflation” which has been echoed by op-ed writers at the Financial Times, albeit with less colorful phrase-making.

    Another reason for the gap between the reported unemployment figures and the lack of wage pressures may be the low threshold for “employment”, which includes doing any paid work or doing 15 hours of unpaid work for a family business. With the large number of self-employed people in the US (and the boundary between self employment and unemployment in fine indeed, since 90% of all new businesses fail in five years and the founder typically is slow to fold the enterprise) we may have more seriously underemployed people here who are nevertheless counted exactly the same as fully employed people.

  9. john c. halasz

    This is a favorite topic of French blogger Laurent Guerby, ( who, note, is not an economist, but an electronics or computer engineer), who comments from time to time on English econ blogs. For example, he notes the unemployment rate for men 25-54 was 4.6% in the U.S. and 7.4% in France in 2004, but the employment to population rate for the cohort was 86.3% and 86.7% respectively. (For the under 25 cohort, though problems and tensions over employment issues are obvious, France has a much higher school enrollment rate). There are also problems with response rates by the unemployed in surveys, which are lower for the BLS than for the U.S. census, for example. And you can add in factors like the high U.S. incarceration rate, which both decreases unemployment and increases employment via the “broken window” fallacy. I don’t know how Ritholtz derived his 6.5% estimate, but there is good reason for focusing on employment to population rates in gaging and comparing internationally unemployment rates and the state of labor “markets”.

  10. Anonymous

    Heh heh heh! There’s now a new problem on how to account for incarcerated people.

    Since the govt has cracked down on immigration, farmers are now using inmates in Arizona to help on their farms.

    Oh the irony.

    Btw, the US is first, or around the top of the rankings of industrialized societies for number of people incarcerated with around 2,000,000 inmates.

    Each inmate costs between $50K-$60K/yr to maintain.

    The prison industry is a very lucrative business.

  11. Anonymous

    BLS provides a range of unemployment rates with only the U-3 given headline status. While this rate may have been 4.6% in July, the more inclusive U-6 measure* was 8.6 and 8.3% not seasonally adjusted and adjusted, down only fractionally from year-ago levels of 8.8 and 8.5%.

    Ideally, the U-6 would provide a closer approximation to reality but, with the late 1980s-1994 shift to new methodologies and definitions, including loss of a U-7 measure, it seems more distortions were introduced.

    *[ U-6 Total unemployed, plus all marginally attached workers, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all marginally attached workers.]

Comments are closed.