How Governments Manipulate Statistics Without Lying — Transcript
Full transcript
- 0:00hiring a lot of PhD statisticians in
- 0:02order to create propaganda apart from
- 0:04anything else is just an expensive way
- 0:05of creating propaganda. Look, if I have
- 0:08to buy you off and say create a
- 0:09department of propaganda and we can have
- 0:11low-level political hacks there who get
- 0:13paid $40,000 a year, at least I'm going
- 0:15to save us a lot of money.
- 0:18>> Hi, Justin. This week you released a
- 0:20video about how in August the Commerce
- 0:23Department quietly scrapped its ban on
- 0:25political interference in official
- 0:27statistics. You talked about how
- 0:29official reports can now be delayed,
- 0:32buried, rewarded, or stripped of their
- 0:34uncertainty. Today, I have several
- 0:36follow-up questions from our audience
- 0:38that I would like to ask you. I'm
- 0:39Augusta. I'm a producer here at
- 0:40Platypus.
- 0:42>> I'm Justin Wolfers. Uh, this is the
- 0:44professor is in. Think of this as office
- 0:47hours. Uh, Augusta is going to open the
- 0:49door. Come on in and bring me your
- 0:52questions.
- 0:53Welcome to my office, Augusta.
- 0:56Okay, this was by far the biggest
- 0:59question that we got from the highest
- 1:01number of people in light of this. How
- 1:04can we believe the most recent jobs
- 1:06numbers?
- 1:08>> Austo, I don't know if you you know
- 1:09this, but when you and I started
- 1:11talking, I woke up this morning in the
- 1:13best mood. I was having the best day.
- 1:15Just thinking about this makes me sad.
- 1:18And the reason it makes me sad is cuz my
- 1:20therapist told me anger doesn't help
- 1:22anyone.
- 1:23Um, okay. Breathe through it. We can get
- 1:28to some of the reasons to be sad or mad,
- 1:31but I'm going to give you the reason to
- 1:33be glad. This whole issue is about the
- 1:36Commerce Department. The Commerce
- 1:38Department just rewrote the Commerce
- 1:40Department's rules to say political
- 1:43interference in statistics is okay.
- 1:47It took the set of rules that provided
- 1:49insulation and said you're not allowed
- 1:51to do political interference,
- 1:52inappropriate influence is inappropriate
- 1:54and so on and it just crossed them out.
- 1:55>> Right?
- 1:56>> The commerce department is the parent
- 1:59of the Census Bureau
- 2:02which collects data like housing starts,
- 2:04retail trade, um the American Community
- 2:07Survey,
- 2:09much of our important social data as
- 2:12well uh as well as congressional
- 2:14districts. Um it's also the parent of
- 2:17the Bureau of Economic Analysis which is
- 2:19most important because it collects the
- 2:21GDP statistics and in fact the personal
- 2:24consumption's expenditure deflator which
- 2:26is the measure of inflation that the Fed
- 2:29uses. So the commerce department
- 2:31directive applies to the census bureau
- 2:33and to the bureau of economic analysis.
- 2:35Collectively those two are two very
- 2:37important statistical agencies. The US
- 2:40unlike other countries has many many
- 2:41many statistical agencies. Just by
- 2:44contrast, the Australia Australia has
- 2:46the Australian Bureau of Statistics one.
- 2:50The jobs numbers come from the Bureau of
- 2:53Labor Statistics, which whose parent is
- 2:56the Department of Labor. This directive
- 2:59was a commerce department directive. It
- 3:01flows straight down through commerce to
- 3:03the Census Bureau and the BEA. But it
- 3:06until the Labor Department adopts
- 3:08something similar
- 3:12then it still has its existing set of
- 3:14protections.
- 3:16Now
- 3:19because of that
- 3:21I don't believe there's been any
- 3:22interference in labor statistics and
- 3:25it's not just because of that. I also
- 3:26have friends on the inside and also the
- 3:29numbers make sense and
- 3:33the sort of conspiracy that would be
- 3:35required to just start telling lies
- 3:37about the unemployment rate is very uh
- 3:41it would involve so many people and this
- 3:42administration leaks anytime there's two
- 3:45people in a room.
- 3:46>> Um so I feel like we're in pretty good
- 3:49shape there. I also want to point out
- 3:51one really important thing for our
- 3:54audience for how you think about how
- 3:56statistics get manipulated.
- 3:59The easiest thing to do is think of a
- 4:00conspiracy. A bloke with a big cigar out
- 4:02the corner of his mouth picks up the
- 4:04phone and orders up uh
- 4:08orders up. Hey, could you make the
- 4:10unemployment rate 2.8%.
- 4:12And you just lie. You just go into
- 4:14Microsoft Word and you delete 4.1 and
- 4:17you type 2.8 Right?
- 4:19In the history of statistical
- 4:21manipulation, it's almost never how it
- 4:22happens. And I think it's really
- 4:24important for people to understand that.
- 4:25What you do is you for instance change
- 4:27the definition of unemployment or you uh
- 4:31change the the statistical weights that
- 4:34are used or instead of relying on the
- 4:37household survey or sorry the the firm
- 4:40survey you you rely on the household
- 4:42survey or you change the industry
- 4:45definitions or you only look at citizens
- 4:48or only look at non-citizens or there's
- 4:51so many ways you can make this month's
- 4:53number incomparable with last month's
- 4:55number or you just bury the number, you
- 4:58stop publishing. If they changed the
- 5:00definition of unemployment, we'd see.
- 5:02So, the thing is it actually happens in
- 5:04broad daylight. And that's partly
- 5:06because trying to manipulate statistics
- 5:08in the dark requires a conspiracy too
- 5:12difficult to pull up.
- 5:15Let me look folks in the eye and make
- 5:17you a promise. As soon as I think we
- 5:20can't believe the numbers, I will tell
- 5:22you.
- 5:23I want to point out and I think this is
- 5:26very important.
- 5:27What they did is they changed the rules
- 5:30to allow the Census Bureau to create
- 5:32propaganda in the future. That is
- 5:35different from a claim that Census
- 5:37Bureau data are unreliable. It says look
- 5:40out this could happen.
- 5:43Um,
- 5:44and so far there's nothing that looks
- 5:47and in fact I believe that the career
- 5:49public servants are going to continue
- 5:50doing their job and continue doing good
- 5:52economic measurements and if you get
- 5:54nihilistic about this and this may be
- 5:57what they want then you say we can't
- 5:59believe the Census Bureau statistics
- 6:02then there's no truth and if there's no
- 6:05truth there's no foundation for any. So
- 6:08you've partially answered this question,
- 6:10but I think it's worth posing directly
- 6:12because a lot of people asked this as
- 6:13well, which was in a broader sense, does
- 6:17this mean that we can trust any numbers
- 6:19that are coming out of this
- 6:20administration?
- 6:22>> So far, there's no evidence that any
- 6:24numbers have been messed with out of
- 6:26this administration. One exception. The
- 6:28exception is the report on so-called
- 6:32voter fraud in the 2020 election. We
- 6:35made a video about that. I lived up to
- 6:37my promise to our audience, which is I
- 6:39looked at her in the eye and I said,
- 6:40"This is a lie. This is incoherent. This
- 6:44is more wrong than it is right. This is
- 6:46the result of political interference.
- 6:48This isn't serious." Wow. That ruined my
- 6:50day when they did that because what it
- 6:52did is it destroyed a century, you know,
- 6:54a decadesl long tradition of attachment
- 6:56to and dedication to the truth by people
- 6:59who work so hard to defend the
- 7:01reputation of an organization
- 7:04that will now take decades to put back.
- 7:07But again, we don't live in a world
- 7:09where you get to make up your own facts.
- 7:12I'm angry about this change at the
- 7:15commerce department because it moves us
- 7:17closer to a world in which a trusted
- 7:19source of truth may become less trusted.
- 7:23But so far this is one of the world's
- 7:26leading statistical agencies and with
- 7:28the exception of one report that had
- 7:29ever released it has been absolutely
- 7:31relentless in publishing its best
- 7:34understanding of the truth. And I know
- 7:37that there are people who are so mad at
- 7:39politics right now, all they ever want
- 7:41to do is throw away inconvenient truths.
- 7:43That might help you feel good, and it
- 7:45might help you win a debate at the
- 7:46basketball court, but throwing away our
- 7:48only understanding of the truth will not
- 7:50lead us to a clearer view of the world,
- 7:54better policies, and better economics. I
- 7:57am not asking anyone to be a fool. A
- 7:58fool is someone who says, "Well, the
- 8:00president may have lied in the past, but
- 8:02I'm going to bully him this time." I
- 8:04don't want that naivity. The rule of
- 8:06thumb was if it comes out of the White
- 8:08House and if it comes out of a cabinet
- 8:09secretary or a political appointees
- 8:11mouth. In the past
- 8:15that would have meant the number was
- 8:16true but cherry picked. In the present
- 8:20it's not even true. Like literally
- 8:23making up numbers. Drug prices haven't
- 8:25fallen 600%. I know that's surprising.
- 8:28The US has not had $20 trillion worth of
- 8:31investment under President Trump. uh
- 8:33because that would be nearly all of GDP.
- 8:36Um so those numbers have all gotten to
- 8:40the point of being so relentlessly
- 8:41unreliable, they're not even
- 8:42cherrypicked and true, they're just
- 8:44false.
- 8:45Numbers that came out of an official
- 8:47statistical agency, the Bureau of Labor
- 8:48Statistics, the Census Bureau, the
- 8:51Bureau of Economic Analysis, the most
- 8:53prominent ones, um those numbers have
- 8:56always been true. They're muddying up
- 8:59the information environment to make it
- 9:01hard to know what, if anything, is true.
- 9:04This upsets me because I come from a
- 9:06world in which I think truth is the only
- 9:08way to freedom.
- 9:10Um, I think the stakes can't be
- 9:13overstated. And with this new directive
- 9:16and what that signals for the future is
- 9:18putting us in the same set of tinpot
- 9:20dictators
- 9:22like Russia and Argentina and Venezuela
- 9:27that is a club we don't want to be in
- 9:29and this is a story that has never ever
- 9:32ended well. So if that is the case, if
- 9:37this is a potential step down a very
- 9:39terrible path,
- 9:42what are the chances or what are the
- 9:44incentives for the next administration
- 9:47to reimplement these protections? Is
- 9:49this high on anyone's priority list?
- 9:51>> There's a reasonable chance that the
- 9:53next administration at some point will
- 9:55ask me for some advice on some stuff. My
- 9:57commitment to truth is absolute.
- 10:00I will tell them let's get statistical
- 10:04agencies in the business of truth.
- 10:06Hiring a lot of PhD statisticians in
- 10:08order to create propaganda apart from
- 10:10anything else is just an expensive way
- 10:12of creating propaganda. Look, if I have
- 10:14to buy you off and say create a
- 10:15department of propaganda and we can have
- 10:17low-level political hacks there who get
- 10:19paid $40,000 a year, at least I'm going
- 10:21to save us a lot of money. It's an
- 10:24economist perspective for you. Um, every
- 10:27serious social scientist I know thinks
- 10:30that the value of data
- 10:33is very high.
- 10:35That adding propaganda to data buys you
- 10:39very little propaganda value, but it
- 10:42undermines the value of all the
- 10:43statistics you've collected because
- 10:44instead of knowing you have the truth,
- 10:46now you know there's some TN you have
- 10:48the truth and some TN you have
- 10:49propaganda.
- 10:50There's a very real question is if what
- 10:52we're going to do is allow the
- 10:54statistical agencies to produce
- 10:56propaganda, maybe we should just slash
- 10:57their budgets because the data they're
- 11:00collecting literally has very little
- 11:02value if we no longer know if it's the
- 11:04truth. I don't want to overstate it and
- 11:06say let's get rid of all statistics, but
- 11:08it's not clear that having statistics
- 11:10mixed with propaganda has much value at
- 11:12all. Uh so for sure I'll be in favor of
- 11:15it. Every serious Republican I know
- 11:19serious
- 11:21they might be doing a lot of work there
- 11:24says that knowing the truth is useful
- 11:26for
- 11:28creating useful policy.
- 11:30So the only counterargument is well our
- 11:32guys can use it to our advantage. We'll
- 11:34leave it to the next guys to clean up.
- 11:36And actually, I think this is at a piece
- 11:38of a broader part of the Trump agenda,
- 11:41which is the president through a range
- 11:43of things, whether it's taking um
- 11:46taking power from Congress,
- 11:49um whether it's moving to a world in
- 11:52which the White House lies rather than
- 11:54cherrypicking truths and so on. He buys
- 11:57himself short-term political gain, but
- 12:00he hasn't thought about the future. in
- 12:01the future is, you know, if he appoints
- 12:02deep partisans to the Supreme Court,
- 12:04what do the Democrats have to do? If he
- 12:06blows out the budget without caring
- 12:08about the future so he can spend money
- 12:10on Republican priorities, what should
- 12:12the Democrats do? And if what we get is
- 12:15the next party says, "Well, you guys did
- 12:17all this irresponsible stuff for
- 12:19short-term political gain. Well, we need
- 12:21that gain no less. We're going to do
- 12:23that." Then we end up in a country that
- 12:24basically can't govern itself.
- 12:27And that's why I really think on all of
- 12:31these issues, Congress just needs to be
- 12:32a whole lot more active because if all
- 12:34you're doing is governing for the next
- 12:36news cycle and you've set up the
- 12:38incentives for the other side to govern
- 12:40for the next new cycle, then you'll
- 12:41never do anything like put the budget
- 12:43anywhere back on a sustain sustainable
- 12:46fiscal trajectory again. this style of
- 12:48short- termism
- 12:50which you know essentially you can think
- 12:52about the old rules were uh we're mostly
- 12:55going to tell the truth. We're mostly
- 12:57going to act in our long-term better
- 12:58interests and occasionally we'll put a
- 13:00thumb on the table our thumb on the
- 13:02scale our way and if what you do is you
- 13:04jump on the scale and you don't bother
- 13:05doing any governing in between then you
- 13:08break the system for both sides which
- 13:09means you're breaking for Americans.
- 13:12So this is my last question and this is
- 13:14uh something a few people alluded to. Do
- 13:18you have a rule of thumb or a way for
- 13:20people to assess statistics who are not
- 13:22trained in it? Any way that they should
- 13:24look at these statistics and think about
- 13:26what they really mean because many of us
- 13:29and I'm including myself in that do not
- 13:30have a PhD in economics and and don't
- 13:32know how to really think deeply about
- 13:34this.
- 13:35>> Yeah. Um yeah. So, first of all, let me
- 13:39just describe the foundational story of
- 13:41Platypus Economics. I want to empower
- 13:43you to be able to do this. If what we
- 13:46did was left at all to talking heads on
- 13:48TV, there are a bunch of dopey leftwing
- 13:50talking talking heads and a bunch of
- 13:52right-wing dopey talking heads. And when
- 13:53you watch TV for too long, all you
- 13:56realize is you got a whole lot dumber
- 13:58and half an hour of your life just
- 14:00slipped away.
- 14:02So, the current ways we think about this
- 14:04or communicated this, I think, are quite
- 14:06broken. Um, and so the goal of Platos
- 14:08economics is to teach the world
- 14:09economics. I want to give you the skills
- 14:11to be able to think through these issues
- 14:14for yourself. My theory of the case, we
- 14:16all have different theories, but my
- 14:18theory of the case is the world will be
- 14:20a better place if people have the
- 14:24economic skills to navigate it a little
- 14:27more subtly. So,
- 14:29glib answer, watch platypus economics.
- 14:32Um, let me give a more serious answer.
- 14:34When you're thinking about statistics,
- 14:37the rule of thumb I gave earlier was if
- 14:39it comes out of a statistical agency,
- 14:41and I want to be clear, that's the
- 14:42Bureau of Economic Analysis, the Bureau
- 14:44of Labor Statistics, the Census Bureau,
- 14:47um, for climate staff, Noah, um, these
- 14:50are official statistical agencies, the
- 14:52Federal Reserve, the Congressional
- 14:53Budget Office, they all tell the truth.
- 14:56if it comes out of someone who if you've
- 14:59ever visited Washington on a school
- 15:01trip, uh the White House is at 1600
- 15:04Pennsylvania Avenue. There's a big fence
- 15:05around it and inside that fence there's
- 15:07a whole lot of other organizations and
- 15:09acronyms. Um there's obviously the
- 15:11president, the vice president, there's
- 15:12also the cabinet. There's the Office of
- 15:14Management and Budget, OM, there's the
- 15:17Council of Economic Advisers, CA,
- 15:19there's the National Economic Council,
- 15:21NEC. These all sound very official, but
- 15:23the important thing to understand is
- 15:24their address is 1600 Pennsylvania
- 15:26Avenue. 1600 Pennsylvania Avenue is a
- 15:29political organization devoted to
- 15:31achieving political ends and at the
- 15:34moment despite the truth that in fact at
- 15:37war with the truth. Um, so recognizing
- 15:40what comes from 1600 Pennsylvania Avenue
- 15:42versus the official statistical
- 15:44agencies. And it gets hard because when
- 15:45we're talking about the budget, a lot of
- 15:47the budget numbers come from the Office
- 15:48of Management Budget and Budget,
- 15:51which is full of lying liars who lie.
- 15:53But a lot of the better projections come
- 15:55from the Congressional Budget Office,
- 15:56which in fact is independent and totally
- 15:59serious.
- 16:01Boy, who can keep track of OM versus
- 16:03CBO? it feels like there's a lot going
- 16:05on and everything's got a a TLA, a
- 16:08three-letter acronym. Um, and so this
- 16:11all becomes difficult. The other thing
- 16:12then is to suggest that you become a
- 16:14discerning consumer of economic news. So
- 16:18this is not an advertisement for me off
- 16:20a platus, but it is thinking about who
- 16:22has a track record of relentlessly
- 16:24telling the truth. A very simple rule of
- 16:27thumb is that someone who's lied to you
- 16:28once will lie to you again. And so in
- 16:31academic life, if you're guilty of
- 16:33intellectual misconduct, of lying, we we
- 16:37drum you out. We say you're not welcome
- 16:39here anymore. And that, I think, is
- 16:42actually useful guide for thinking about
- 16:44all of these things. So there are many
- 16:46so-called experts that you'll see um
- 16:49turn up on cable TV, Larry Cuddlo,
- 16:52Steven Moore, Peter Navaro, who are
- 16:55lying, liars who lie. And once they've
- 16:57lied to you, once they will lie to you
- 16:58again. get rid of them. They're a very
- 17:01good and very serious conservative
- 17:03economist. Greg Mankey is a very dear
- 17:04friend of mine. Doug Holtzen has
- 17:06enormous integrity. Um they're a very
- 17:09good and very serious leftwing
- 17:10economist. Um so be very thoughtful in
- 17:14that diet. Um, and then anything that
- 17:16has a name like America First or,
- 17:20you know, America's Super Dupart, you
- 17:22know, all these think tanks and groups,
- 17:25they all seem to be called America
- 17:27Eagles, Fireworks, Guns, Woohoo, um,
- 17:30they're not very imaginatively named,
- 17:32but if they sort of sound like
- 17:35um, a bad college rock band, um, they're
- 17:39probably in the propaganda business.
- 17:41Again, not uniformly true. They're very
- 17:42good and very serious think tanks. The
- 17:44Heritage Foundation is a joke that
- 17:45hasn't told the truth in decades. The
- 17:47American Enterprise Institute is a
- 17:50centerright organization that I held in
- 17:52quite high esteem. The Brookings
- 17:54Institution is a very serious outfit
- 17:55that I hold in very high esteem closer
- 17:57to center left. Um, so thinking about
- 18:00the WHO, unfortunately, is a very easy
- 18:03shortcut to thinking about whether
- 18:05something is likely to be truthful.
- 18:08>> Thank you, Justin. And thank you so much
- 18:10to our audience for these questions. We
- 18:12really hope that you enjoyed and learned
- 18:14something from this episode of The
- 18:16Professor is In. If you would like your
- 18:18question answered in a future segment,
- 18:20leave a comment if you're watching this
- 18:22or listening to this. And make sure to
- 18:24like and subscribe to Justin and
- 18:26Tupatopus Economics on YouTube, on
- 18:29Substack for an audio only format. You
- 18:32can subscribe to us wherever you listen
- 18:33to your podcasts. Thank you so much.
- 18:36>> And one last thought before we go.
- 18:39Washington won't care about this issue
- 18:41or any issue unless voters care about
- 18:44it. The whole point of trying to
- 18:46describe all of this is so that folks
- 18:47out there can understand what's going
- 18:49on. And if you get active, then we can
- 18:51get a step closer to the truth. Uh,
- 18:54look, none of us know exactly what the
- 18:56truth is, which is why I always end on
- 18:59stay curious.
About this transcript
This page contains the full transcript of How Governments Manipulate Statistics Without Lying by Platypus Economics with Justin Wolfers, generated from the public captions YouTube serves with the video. The transcript has 3,295 words across 503 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.