Guest Contribution: “Trump Jr.’s Pants-on-Fire Allegation of Manipulated Jobs Numbers”

Today, we present a guest post written by Jeffrey Frankel, Harpel Professor at Harvard’s Kennedy School of Government, and formerly a member of the White House Council of Economic Advisers.

When interviewed about the unemployment numbers, which have fallen steadily since 2010, Donald Trump Jr., replied “These are artificial numbers. These are numbers that are massaged to make the existing economy look good, to make this administration look good when, in fact, it’s a total disaster.” PolitiFact asked a variety of experts about the quote. Their bottom line: the quote from the younger Trump was a “Pants on Fire” lie. The truth is that presidents don’t and can’t manipulate the jobs numbers. No White House has even tried — at least not since Richard Nixon made a heavy-handed attempt in 1971 to interfere with BLS staffing. After that, extra firewalls were put in place.

Here is my own full response to PolitiFact’s question regarding the Trump claim:

The statement is 100% false. The employment numbers come from the Bureau of Labor Statistics (part of the Labor Department). In this administration, like every administration, those who produce the employment statistics are long-time nonpolitical professionals. The Secretary of Labor does not even know what the numbers are going to be when they are announced every month (the morning of the first Friday of the month).

Allegations that the official government numbers understate unemployment are sometimes based on a claim that some higher measure (which, for example, includes discouraged workers who have given up looking for a job, or part-time workers), should be used in place of the ones that get the most attention in the press. But these other measures are also made publicly available by the BLS and the press is free to write about them as much as they want.

The important thing, of course, is to be consistent across time in which measure you use. It wouldn’t be right to switch from looking at the conventional rate to a measure that includes discouraged workers just because you don’t like the incumbent president and want to make things look bad for him.

49 thoughts on “Guest Contribution: “Trump Jr.’s Pants-on-Fire Allegation of Manipulated Jobs Numbers”

  1. Paul Mathis

    Two More Big Lies

    “Wages have stagnated for 35 years.”

    In fact, Real Median Weekly Wages are now at an all time high and rising. They are well above the level of 1981.

    “NAFTA and Pres. Clinton destroyed manufacturing jobs in America.”

    Manufacturing employment INCREASED under Pres. Clinton after NAFTA was ratified in 1993.
    Under Pres. George W. Bush, manufacturing employment dropped by 27%, the absolute worst record by far since the end of WW2.
    The GOP, not NAFTA, killed manufacturing employment in the U.S.

        1. Fred Fnord

          1) Median Weekly Wages measure both number of hours worked and the pay per hour. So if everyone was suddenly required to work 80 hours a week, then median weekly wages would double. This is not ordinarily considered to be that great a thing.

          2) What’s your definition of ‘stagnant’? Clearly you are delighted by an average increase of 0.5% annually (even discounting issue #1), so would you still be talking about how great this increase was if it were 0.2%? 0.1%?

          3) ALL of the weekly gains since 1981 have been from women gradually increasing their work hours outside the home. If you look at the graph of men’s weekly wages, here:

          Then you see that men’s income is completely stagnant, except of course that it was much higher in 1979. Whereas women’s gradually increases, both because women are earning more per hour than they used to and because more women are working more hours.

          4) You’re rather arrogant for someone who doesn’t know what he’s talking about, aren’t you?

          1. baffling

            fred, why should you expect median real wages to rise? should flipping a burger at mcdonalds demand a higher real wage today than ten years ago?

        2. Alex

          Gee, I don’t know Paul.

          Here is what FRED said:

          Regarding that chart:

          Employed full time: Median usual weekly real earnings: Wage and salary workers: 16 years and over

          CPI Inflation Calculator from the U.S. Bureau of Labor Statistics

          Has the same buying power as:
          $335.00 in 1979 equals $1,112.23 in 2016
          $346.00 in 2016 equals $346 in 2016


          Here’s the Contact page:

          Go ahead and reach out to them. Maybe you can impart your wisdom to them.

          1. Paul Mathis

            Good lord!

            The $312 REAL Median Weekly Wage for 1981 which I referenced is based on this:
            1982-84 CPI Adjusted Dollars,
            Seasonally Adjusted”

            That information is taken DIRECTLY from the graph. But then for some unknown reason you adjusted that to 2016 dollars when the 2016 rate has already been adjusted to the REAL value based on 1982-84 dollars. Did you graduate high school Alex?

          2. Alex

            Well Paul, as long ago as that was, I do remember they didn’t teach economics in high school.

            And even in college, a typical student may take a couple of econ classes.

            The original comment I made was about wages being stagnant.

            So, let’s look at what some people who are economists have said about this:

            FRED breaks it down between sexes with Median usual weekly real earnings: since 1980, men’s wages have been flat and women’s have risen slightly.

            Pew Research shows that since 1964, in 2014 dollars an increase from $19.18 to $20.67

            EPI – Again, broken down by race, it shows flat or drops in wages from 2007 – 2014.

            Again, we read that if wages had kept up, the minimum wage would be $9.38 in 2015 instead of $7.25.

            There’s plenty of evidence out there from reputable professional sources to support that wages have been stagnant, and yet you ignore it with your posturing.

            Trump Jr has a point when he said there’s a disconnect between Washington and the rest of the country.

            People wonder why the Democrats are losing voters when someone like Sanders shows up. Wonder no more. Dems stopped listening to their base of working people a long time ago.

            Btw, is this your unreviewed self-published book?

        3. Tim Keating

          But that’s only looking at one axis of the measure. US GDP has gone from $3.2 trillion in 1981 to $17.9 trillion in 2015, an increase of about 5.5x. But you have to factor in population growth, because some of that is doubtless from throwing more bodies at the economy. Population grew about 40% between 1981 and now, so divide your growth of 5.5 by 1.4 and get a per-capita net GDP gain of roughly 3900%. Of that, the average worker has recognized 11%.

          So while you’re right in saying that wages have not literally stagnated, when you compare wage growth to economic growth, there’s a massive disparity. In real inflation-adjusted terms, labor is doing about 10% better than they used to be, but they ought to be doing better.

      1. Charles

        And during the Obama years? ….. since you make it a habit of focusing on the 8 years of Bush.

        Let’s see, real weekly media wages rose from $345 in Q1: 2009 to $346 in Q2: 2016.

        Whoa! Could you bring me an extra six-pack of baloney… go with your Obama bull — $1 increase over 7 years, 14 cent/year! It just don’t get any better for middle class working folks than this…….

        1. Paul Mathis

          @ Charles

          Yeah, that is really interesting. Wages went from $331 in 3Q08 to $345 in 1Q09, an annualized growth rate of 8.5% when they had been declining over the previous 7 quarters. I wonder what Bush did to produce that massive increase in wages?

          Oh right, there was a massive economic collapse happening and millions of lower wage workers were losing their jobs as the economy plunged into the Great Recession. So due to a statistical quirk over 2 quarters, Obama’s record looks weak. Thanks for pointing that out Charles. You are really insightful!

        2. mx9000v2.8

          Wow, how does an economist forget the Great Depression of 2007, with hundreds of thousands of job losses under Bush. And then you expect wages to go up, with massive Bush unemployment, that needs to be undone? No sane person would expect wages to go up until those employment numbers improved. Which they have for 7 years.

          1. PeakTrader

            I stated before, in 2009, we needed a bigger tax cut (to refund consumers), a massive overhaul of regulation and litigation (rather than piling more), massive changes in the tax code, and raising the minimum wage.

    1. PeakTrader

      From 2000-07, U.S. manufacturing productivity had record breaking gains, rising 4.7% a year on average.

      1. Paul Mathis

        And yet EVERY YEAR during Dubya’s presidency manufacturing employment declined. Truly an amazing record unmatched by any other president in the post-WW2 era.

          1. Paul Mathis

            @ Peak Trader
            Of course, I love to see manufacturing workers lose their jobs to automation because who in their right mind wants to work in a factory when they can be sitting home collecting unemployment benefits and watching TV?

            Even Trump understands this situation better than you do.

          2. baffling

            peak, on a macro scale, producing cheaper with fewer inputs does not help if in the process you shrink the consumer class.

          3. PeakTrader

            Paul, again, you don’t know what you’re talking about.

            Producing more output creates jobs, which a dynamic economy creates.

          4. PeakTrader

            Baffling, lower production costs and lower prices free-up limited resources for the economy to expand and give consumers more goods.

    2. Ronald Calitri

      Really? Your FRED figure shows an increase in median wage from 335 to 345 chain linked dollars. That’s ten dollars over 45 years, two bits and a couple of pennies.

      Generally, Trump’s sarcasm can be trusted to point at an issue, if not address it. Prof. Frankel protests way too much in defense of a path-dependent method of survey tabulation. EEven non-labor economists are aware of this, and wish for a method of counting unemployment that would induce correction. That one or another algorithm has been waved ritually for generations is not necessarily a mark of effectiveness.

  2. spencer

    Paul, I agree.

    The dominate problem with US industry has been China. But too many people who should know better blame NAFTA. I do not understand how they get away with it and it is both the left and the right that do it.

  3. baffling

    the focus on a crappy economy is simply conservatives rewriting history to tell a story that benefits them politically. but it is typical behavior. they have solutions to problems, and if those problems do not exist in reality, promote an alternative reality where those solutions are grand.

    NAFTA is one of the interesting issues. from a conservative viewpoint, i would think free trade is a pinnacle. but interestingly, conservatives like to bombard clinton for approving NAFTA, but as I recall it was george hw bush who negotiated most of the treaty. again, simply rewriting history for convenience.

  4. Alex

    What does the data say?

    I wrote about this in March (links to BLS data in the article):

    Under Clinton in 1994, we see two new categories:

    We see a trend with DISCOURAGED WORKERS averaged by term:
    Clinton: 359K
    Bush: 408K
    Obama (thru 2015): 873K

    We see a trend with MARGINALLY ATTACHED WORKERS averaged by term:
    Clinton: 8.8%
    Bush: 9.2%
    Obama (thru 2015): 14.3%

    So with the new buckets created by Clinton and Rubin, where are they drawing from? A strong case can be made that people that were counted in the regular unemployment numbers are not being counted.

    From 1977, we can check PEOPLE NOT IN LABOR FORCE, which was fairly consistent under Carter and Reagan, but then jumps significantly:
    Carter: 1,051,000
    Reagan: 1,702,000
    Bush Sr: 2,976,000
    Clinton: 4,926,000
    Bush Jr: 10,292,000
    Obama: 13,533,000

    In addition, we find Obama ordered the BLS to stop counting mass layoffs in March 2013:

    We can also look at LABOR FORCE PARTICIPATION RATES averages by term:
    Carter: 63.2%
    Reagan: 64.7%
    Bush Sr: 66.4%
    Clinton: 66.8%
    Bush Jr: 66.2%
    Obama (thru 2015): 63.8%

    The other thing no one catches is the monthly Jobs Created revisions. Every month the numbers issued are preliminary numbers indicated by “(P)”:

    May 2016 numbers were originally announced as 38,000, which is bad. They have since been revised downward to 11,000 (which is very bad.

    Not a peep by the mainstream media on that.

    Last, they tell everyone that the U-3 unemployment rate as of June is 4.9%, which is the range of full employment. It’s completely unbelievable. Especially when there are 94,517,000 NOT IN THE LABOR FORCE:

    Btw, I am not voting for Trump or Clinton. I may sit this one out or vote for my dog.

    1. Jeff Frankel

      I take exception to three of your points.
      (1) Yes, of course if you start counting from January 2009, then the average level over Obama’s term does not look so good. At that moment, when he took office, the economy was in the worst free fall since the 1930s. We already know that. But take a look at the graphs. The rate of job loss and the other measures began to turn around rather quickly after he took office. Employment change then turned strongly positive.
      (2) On the subject of the unusually low jobs number in May and its revision further downward, you say “Not a peep by the mainstream media on that.” To the contrary, this was reported in the media. Personally I think there is too news space devoted to one-month numbers, which are very noisy as everyone knows. The result is that they receive little attention (outside of the financial markets). I think this is because the noise to signal ratio is too high. In my view it would be much better to focus on the most recent 12-month change in employment (as some other countries do) rather than the most recent 1-month change. But of course this is a comment on what the public wants to hear, i.e., short-term “breaking news.” Anyone can choose to focus on the 12-month change, which is what I always recommend. (Job creation looks very good for Obama over the last 5 years, except for that one single month.)
      (3) “they tell everyone that the U-3 unemployment rate as of June is 4.9%, which is the range of full employment. It’s completely unbelievable. Especially when there are 94,517,000 NOT IN THE LABOR FORCE” Surely you know that the denominator of the U-3 unemployment rate is the labor force. Your point about those not in the labor force is a completely irrelevant in this context.

      1. Alex


        I’ll give you point 2, but I’m still holding on 1 & 3.

        I only usually concede after a couple of bottles of Guinness Draught on a sunny day. Today, I’m on my second bottle and it’s sunny outside in the 80s.

        Here’s an idea for you for a future article:
        How much have wars (and all defense-related spending) AND tax breaks (subsidies, etc) cost U.S. taxpayers since 2001?

        I came up with a ridiculous number:

        When I look at the number of wars the U.S. has been in since 1775, it’s amazing. By my count, Obama has had 15 during his time in office (the most of any president, followed by Pierce, Buchanan, and Lincoln):

        The U.S. has been at war with someone somewhere in 213 of 242 years since 1775 (88% of its history).

        I’m curious to see what you come up with.

  5. Steven Kopits

    “long-time nonpolitical professionals”

    My wife is active in labor markets issues in New Jersey. I don’t believe I have ever met a Republican involved in labor markets from the government or academic side.

    I would be impressed if you could find me a single registered Republican analyst or manager at BLS dealing with labor statistics. I would be astounded if you found me two of them.

    Thus, BLS statisticians may be ‘longtime professionals’, but I think it probably fair to assume that the staff at BLS are overwhelmingly Democrats with some portion of independents and very, very few Republicans. Whether this influences analysis is a matter of organizational culture.

    1. Menzie Chinn Post author

      Steve Kopits: I can actually speak to this issue having spent time at CBO, a couple Feds, and the Board. I usually cannot tell from what the individual says with respect to the economics or the statistics what their politics are. Sometimes, I find out, and am surprised.

      So if you’re working off anecdotes, well there you have one.

      And if you ever bump into Doug Holtz-Eakin, then you will have increased the number of academic labor economists that are Republican by an infinite percentage…(Then you’ll have *two* anecdotes to bandy around!!!)

      1. Rick Stryker


        Let’s not forget Casey Mulligan and one of your favorite economists, Ed Lazear.

        1. Menzie Chinn Post author

          Steven Kopits: OK, I worked at the Executive Office of the President, too. Professional OMB staffers — I couldn’t tell what their politics were from what they said. People I met from State, Treasury, Energy, other agencies, I couldn’t tell. I think I’ve got more anecdotes than you, here…

          1. Steven Kopits

            I think labor is different, Menzie. Macro, trade, finance, these will draw people from all sides of the political spectrum. In my experience–albeit that may be colored by my Princeton and New Jersey experience–the labor guys are all on the left.

    2. baffling

      or perhaps steven, it is simply hard to find a republican who practices nonpolitical labor economics?

      1. Steven Kopits

        Baffs –

        I did not state that Republicans practice non-political labor economics. However, I did suggest that a preponderance of ideological orientation in an institution, regardless of whether it is right or left, will tend to produce results that favor that ideology.

        I was arguing that the phrase “long-time nonpolitical professionals” is asserted, not proven. I think Republicans can assume a vast preponderance of non-Republicans at the BLS. Whether or not this influences outcomes there, I don’t know. At the DOE, I can certainly see the influences from time to time, and at EPA, it’s just ridiculous. At BLS, I don’t know.

        1. baffling

          steve, as an anecdote, i have a cousin, republican, who works in the dept of labor. i never hear him complain about the politics of the department.

          1. John

            BLS is left wing, DOD is right wing. No secret, & exceptions confirm the rule.

            One could speculate, why HRC will be eager to start a war after the MSM has put her into the presidency. My guess is, her DOD friends want more influence and some cash-back.

  6. Rick Stryker

    Time to fact check the fact checkers once again. In this installment, we fact check the Politifact claim that Donald Trump Jr. committed a “pants on fire” lie when he allegedly said that government unemployment statistics were manipulated for political purposes.

    To judge Politifact’s claim, we look at the video of the inteview and the transcript. An examination of the transcript record shows that Trump Jr. nowhere said that unemployment statistics were manipulated for political purposes. Thus, Politifact’s allegation must rest on their view that Trump Jr’s use of the terms “artificial” and “massaged” implied that he meant these terms to indicate manipulation for political purposes.

    To judge whether Trump Jr. actually meant to imply political manipulation by using those terms, it is necessary to look at the general context of his remarks, something that Politifact failed to do. Here is the full context:

    TRUMP JR: “…Let’s talk about the real unemployment rate. Because the way we actually measure unemployment is after X number of months if someone can’t find a job, congratulations, they’re miraculously off.

    When you talk about underemployment which Obamacare has destroyed, people that are working 30 hours a week instead of 40 hours a week so companies don’t have to put them on Obamacare. When you talk about people that just aren’t even registered because they don’t count them anymore. They have been out of work for so long. They’d love to work if they could but they can’t that doesn’t count.

    These are artificial numbers, Jake. These are numbers that are massaged to make the existing economy look good and make the administration look good when in fact it’s a total disaster.

    So let’s talk about real numbers. You know, when we talk about the numbers as they see it it’s fake accounting, it doesn’t really work. And that’s not what the actual state of this country is.”

    Trump Jr. appears to be using the terms “artificial,” “massaged,” and “fake accounting” to mean misleading. Is this how Jake Tapper understood Trump Jr? Or did Tapper say, “Wait a minute Mr. Trump. Are you alleging that the Administration is cooking the numbers?” Let’s look at the transcript.

    TAPPER: “So when your father – if your father becomes president he won’t use the Bureau of Labor Statistics unemployment rate. He will use the overall rate which includes people who have left the workforce and people who work part time. He’ll – he’s going to switch because every president uses – has used the typical number. He’s going to use this number that includes everything?”

    Thus, Tapper understood Trump to be saying that the standard unemployment rate is misleading in the current situation. He asked if Trump’s father would use a different unemployment series that better counted the underemployed if he became President. Did Trump Jr correct Tapper, saying, “no, you misunderstood. I was saying that the numbers are dishonestly manipulated and that won’t happen in my father’s Administration.” Or did he accept and agree with Tapper’s interpretation of his remarks? We look at the transcript.

    TRUMP JR.: “Listen, I don’t know what he’s going to do, I’m not a policy wonk in terms of what we are going to decide to do, but I think we have to acknowledge that there’s those numbers and then there is the real situation and they are very different. OK?

    We all understand that, you are acknowledging that at least. Let’s talk about it. I think those are the people that my father is speaking. Those are the people that have been forgotten in this country. And that’s exactly what it is.

    You know, unemployment is great because these people just can’t even find jobs so they don’t count anymore. Those are the people we’re talking to. Those are the people we want to put the work, Jake. Those are the people that can’t feed their families and we’re telling them that they’re employed because some bureaucrat in DC says, well, this is the way the numbers work. It doesn’t work that way in the real world.”

    Trump Jr. confirmed Tapper’s understanding that he meant the standard unemployment rate is misleading because it does not adequately count people who have dropped out of the labor force. We thus rate Politifact’s claim that Trump Jr was implying that the government is manipulating employment statistics to be FALSE.

    What of the substance of Trump Jr’s remarks? This chart shows the relation between the standard unemployment rate and U-6, a version of the unemployment rate that has a broader definition of the unemployed, in line with Trump Jr’s point. (It includes those working part time for economic reasons for example.) Comparing comparable times in history when the unemployment rate was about 5%, we see that over the second half of 2005 U-6 is about 4 percentage points higher than the conventional unemployment rate while recently U-6 is about 5 percentage points higher. Trump Jr’s point is also buttressed by a look at the employment to population ratio, which has not recovered to pre-recession levels.

    Since we have determined that Politifact’s principal claim is FALSE and that there is empirical support for the point that Trump Jr. was trying to make but which Politifact obfuscated, we award 3 Strykes to Politifact for mendacity. We also award 3 Strykes to econbrowser for repeating these false claims.

    Rating system:
    1 Stryke: reasonable people can disagree
    2 Strykes: fact checker has finger on the scale in favor of the Democrats
    3 Strykes: in the tank for Hillary
    4 Strykes: part of the DNC conspiracy to cheat Bernie out of the nomination

    A note on FactCheckBusters:

    FactCheckBusters, located on the beautiful Wossamotta University campus, was founded by Prof Rick Stryker upon the realization that the media fact checkers are not objective arbiters but rather thinly disguised joint ventures with the Democrat-supporting editorial pages of the major newspapers. From time to time during the Presidential campaign, we will restore balance by fact checking the fact checkers.

    1. steve

      Sorry, your comments don’t pass the sniff test. Trump Jr knew exactly what he was doing, and he was making it clear that numbers are being manipulated to benefit Obama. Nice try.


      1. Rick Stryker


        I’m not surprised that you feel that way but then I didn’t write this comment for you. Partisans such as yourself, 2slugs, Menzie, etc and others will always hear the “dog whistle,” because it’s in their interest to hear the dog whistle. I understand that the actual facts don’t matter to you. Attacking positions that conservatives don’t really have is a standard tactic of the Left. It’s much easier to attack a point that Trump Jr didn’t really make than it is to deal with his real point, which is that the employment market is actually worse than the standard measures would indicate. I don’t expect you guys to change.

  7. 2slugbaits

    Rick Stryker Once again you have managed to mangle the English language. The dictionary defines “massaged” as a transitive verb (go look it up). That means there has to be a subject acting upon a direct object through the transitive verb. So when Junior talked about the employment numbers being massaged, the only way to make that a proper sentence is to assign a subject and a direct object; e.g., “The Obama Administration massaged the employment numbers.” You’re trying to make the verb “massaged” into an intransitive form, which it isn’t.

    If Junior intended to make the relatively innocuous claim you suggested; viz., ..the standard unemployment rate is misleading because it does not adequately count people who have dropped out of the labor force.” then why didn’t he say exactly that? Here’s a theory. Because Trump and his apparatchiks prefer to communicate in dog whistles. Junior knew perfectly well that the typical low information voter was hearing something along the lines of “Crooked Barack’s folks are cooking the books to make him look better.” That’s why he blathered on about “artificial” and “fake accounting” and “massaged.” Those are all loaded words meant to convey evil intent. It may well be the case that the headline unemployment numbers are misleading in the sense that they don’t tell the whole story, but that’s hardly a problem unique to the Obama years.

    Helpful hint for future blogging. If you want to have any credibility, then resist the temptation to mangle the English language beyond recognition. That’s why people don’t trust lawyers.

    1. Rick Stryker


      If someone plainly says something, fine. Or if you can reasonably infer someone’s meaning from context, fine. But if you have to resort to hermeneutical analysis of the parts of speech or listening for “dog whistles” to infer someone’s meaning, then you are in danger of hearing what you expected to hear rather than what someone is actually saying.

      Jake Tapper didn’t hear the dog whistle. He heard what Trump Jr was actually saying and asked a very relevant question: would Trump Sr. use a different unemployment series if he became President, one that better accounted for underemployed people? Consistency would suggest he should but Trump Jr. dodged that question of course.

  8. Joseph

    2slugbaits, go easy on Rick. He’s so naive that he believes that people in ninja costumes are riding jet skis across lake Erie to invade the Rock N Roll Hall of Fame in Cleveland.

    1. Rick Stryker

      See Steve, this is what I was talking about.

      Joseph is trying to discredit me by attributing a ridiculous statement to me. But of course I never said that.

      I regret that I will have to award 2 Strykes to Joseph. Obviously FactCheckBusters is going to be very busy during this campaign season. I’d better beef up the staff!

  9. Paleo 6

    Semantics and statistics-2 good ways to spin and distort data

    So I ask- when wage gain analysis is shown, is it inflation adjusted?

    Yes, the Unemployment rate is maybe 4.9% But what about the % of people who don’t have a job?
    If you calculate the People out of work statistic, the way it was calculated during Truman’s or Eisenhower’s time, it is now 23%.

    Abetter way to look at these numbers, is Number of People Employed! Participation I believe its called
    And every year since 2008 that number has gone done. Now ~ 66 million or so
    the lowest number since Carter. And the US population is much greatere

    1. baffling

      and the large baby boomer population is retiring, ie leaving the work force. you have a larger us population, but a greater percentage of that population is retiring compared to the carter years in your example. you need to consider demographics in your argument.

Comments are closed.