Latest / Words & Numbers / Episode 446: The Job Numbers Shock

Transcript

Episode 446: The Job Numbers Shock · Words & Numbers

0:00
Hi, I'm James and I'm Anthony and this.
0:02
Is words and numbers. So what are we going to talk
0:09
about this time around, Aunt? This week on Words and Numbers,
0:12
I want to talk about denaturalizing citizens.
0:15
Would you believe the government is actually talking about
0:18
stripping Americans of their citizenship?
0:21
Well, imagine that. Imagine that.
0:24
It's not like they've never done that before.
0:26
But you can see under the current climate how this might
0:29
be a little more threatening, a little more ominous.
0:32
So we'll see where that goes. But you got to stop saying this
0:36
week we go twice a week now. Man, I want to get into it.
0:38
I mean, it's weird because you want to talk about something
0:41
political. I want to talk about something
0:42
economic. There's a very strange set of
0:46
economic circumstances happening right now that only happened
0:51
once before and it was in 1987, and you remember what happened
0:55
in 1987. 01987 was the stock market crash because nobody
1:00
remembers it was a. A A very, very big crash didn't
1:05
last for long for whatever reason.
1:07
But no, it didn't. But we're repeating the intro to
1:11
that, so we'll see if we repeat the outro as well, which would
1:15
be, I think, unfortunate. That, of course, will bring us
1:18
to the foolishness of the week. And I don't even know how to
1:21
explain in polite language what I'm going to talk about.
1:26
But let's just say people need to stop throwing things onto the
1:30
court at WNBA games. I'm interested to see how you
1:33
managed to do that politely. On to the main episode.
1:36
I want to talk about the kerfuffle at the Bureau of Labor
1:40
Statistics. What did they do?
1:42
What are we going to do about it?
1:44
Should we do anything about it? Well, they fired that nice lady
1:47
who gave them. I mean, it's just a classic
1:49
shoot the messenger issue, right?
1:51
The messenger was killed here because of bad jobs numbers.
1:55
And now she's gone and Donald Trump will put somebody else in
2:00
there. He's already nominated somebody,
2:01
by the way, who is absolutely crazy.
2:04
We'll get into that some other time.
2:06
He will nominate somebody who will give him numbers he wants.
2:09
Yeah. But the the Bureau of Labor
2:11
Statistics, long the gold standard of government stats,
2:16
will now be suspect. Yeah.
2:18
And this is one of the things we want to look at because the
2:20
thing that we're told on one side is that BLS is getting the
2:24
numbers wrong because there's something fundamentally flawed
2:28
at the BLS. The story on the other side is
2:31
that the administration just simply doesn't like the answer.
2:34
All this and more of this week on words and numbers.
2:37
See Eve and I did it this time on words and numbers, which Ant,
2:41
by the way, is a civil original podcast.
2:44
Check it out at civil.com C ivl.com.
2:49
So what's new and exciting in your world this weekend?
2:52
James, have you ever made an error on your taxes?
2:59
Yes, Yeah, Yeah, I did. I have and.
3:04
Curious thing though if if I may, when I make an error the
3:07
IRS calls me on the phone and tells me I made an error.
3:10
It tells you I made an error. I know exactly right, right.
3:12
I, I made an error. This was early on, like I was
3:15
still in Graduate School at a time and, and anyway, it enough
3:21
stuff fell on my head that I said never again from that point
3:25
forward. I from that point forward I was
3:27
still filing A-1040EZ but I get an accountant to do it.
3:31
Never again am I do I want to deal with that.
3:33
Nonsense, as God and nature intended, but.
3:37
Here's the thing. The Department of Justice has
3:42
established de naturalization as one of its, quote 5 enforcement
3:48
priorities. That is people who run afoul of
3:53
the law, the department in certain ways, the Department of
3:56
Justice is now looking to de naturalize them.
3:59
That is, there are people who weren't native born in the
4:03
United States, became citizens and now we're going to strip
4:06
them of their citizenship. And one of the things now that
4:10
you can be de naturalized for is, quote, financial fraud,
4:15
which I should say financial fraud against the United States,
4:19
which explicitly includes tax violations.
4:24
Yeah, you can see all the ways this is going to go horribly,
4:27
horribly wrong. Oh.
4:28
My God. But don't don't worry, because
4:33
this is going to be tied up in court for the rest of the Trump
4:37
administration, and whoever comes next will silently drop
4:41
this. You think so?
4:43
This is a terrible idea. Well, I think it's a terrible
4:46
idea, too. Here's I'm you're, you're giving
4:49
me some solace. But here's my concern.
4:52
First off, I didn't know that denaturalization was a thing.
4:55
I mean beyond the obvious treason for.
4:58
Example. I mean it, it is, but it isn't
5:00
really well. Here's where I would warn people
5:04
and, and I'll take you on your word here that we don't need to
5:07
be worried about this, but, but I want to put a bug in
5:10
everybody's ear should this come up again.
5:13
Look, they're talking about stripping people who have become
5:17
citizens of the United States, stripping them of their
5:20
citizenship. To short step well.
5:24
Well, you can't really strip American native born Americans
5:28
of their citizenship because where would they go?
5:31
Unless, and this is AI guess, a recent change in the law of the
5:36
United States now recognizes dual nationals.
5:39
So it's just one step from that to if I, for example, a native
5:45
born American, but I also hold a citizenship in some other
5:50
country, why can't they strip me of my US citizenship and just
5:54
send me to that other country? Yeah, I, I think what it reminds
5:57
me of is what people might not know this about me.
6:00
I used to be a DJ at a dance club.
6:02
And, you know, at the end of the night, the DJ always says the
6:05
same thing. It's closing time.
6:07
We don't care where you go, but you can't stay here.
6:12
And you know, that's the danger that you get kicked out, but you
6:17
don't have anywhere to go. And when that happens, you just
6:19
go to prison. But I don't think, I don't, I
6:22
don't think this is going to turn into a problem.
6:25
There are a million other things that will, but I'm I'm not
6:27
particularly worried about this one.
6:29
So we'll see. I mean, you never know.
6:31
I've been wrong before on my taxes for example, but there's.
6:36
All kinds of things that a year ago I would have said are not
6:40
likely to happen have been. Happening yeah no I see I used
6:43
to say never in a million years turns out I was off by roughly a
6:47
million years but but Ant I've got I've got something for you
6:52
and this is going to be a little strange by way of how we do
6:55
things here but I'm going to read you something from
6:59
marketwatch.com and this is a a website that you know is does
7:05
reasonably solid work on economic.
7:07
I read MarketWatch. There's nothing.
7:10
These aren't crazy people is what I'm trying to emphasize
7:13
before I get started. And I got like 3:00-ish
7:17
paragraphs for you and I want to get your take on it.
7:20
This past week has been strong for the US Dollar Index.
7:23
It's best since 2022. In fact, it's a far cry from the
7:27
first half of this year, when the index lost almost 11%, the
7:31
worst first half return on record since the index was
7:34
created in the early 1970s. Looking back, you might think
7:39
that the weak dollar was good for U.S. stocks, given the S&P
7:42
500's total return of 6.2% during this period of dollar
7:47
decline. But the US stock market has also
7:50
performed spectacularly in years when the dollar was unusually
7:54
strong. So maybe the dollar's movements
7:56
don't matter to dollar based investors.
7:58
Or just maybe, if the Federal Reserve lowers US interest rates
8:03
aggressively as the Trump administration wants, and the
8:06
dollar declines sharply as a result, the stock market could
8:10
react adversely. What do you think?
8:13
The next sentence is there? Do you have any idea?
8:16
I presume it's now, well, these are reasonable people, so I
8:20
don't know. I was about to say it's going to
8:22
be a call to the Fed to cut interest rates, but I'd be
8:25
surprised if MarketWatch said such a thing, no?
8:27
No, they do no such thing. It's happened before, in October
8:31
1987. For those of you out there who
8:33
are too young to remember October of 1987, there was a
8:37
very grim day that month. Very grim.
8:41
Yeah, it was a stock market crash.
8:43
That was, if I'm remembering correctly, percentage wise it
8:48
was as large as the 1929 crash. I think it was close and only
8:52
second to the 29 crash. So this was a big deal, and we
8:57
may be looking at the same set of circumstances that preceded
9:02
that. Now, notice, I'm not saying that
9:05
these are causal. I don't know if they are.
9:07
If they aren't. And the way this article is
9:09
written, I don't think anybody's making that case.
9:12
It's just something that we should know, we should be
9:15
concerned. Yeah, I, and that's funny
9:17
because you were, you were not overly concerned about
9:20
Department of Justice and denaturalization.
9:22
I'm not overly concerned about this one.
9:24
Here's why we had this crash. It was a 20% drop give or take
9:29
in on one in one day in 1987, October 1987.
9:34
It was because I've looked at the data, it was maybe 3 months,
9:38
possibly six months later, but there was no sign anywhere in
9:42
the economy. Yep.
9:44
Did they never crash? So it was like it never
9:46
happened. And and the difference is that
9:49
in 1929, the Fed then did the exact opposite of what it should
9:54
have done in the face of a crash like that.
9:57
In 1987, the Fed did what it should have, what it should have
10:00
done in 1929, and you had a crash and then the thing
10:04
reverted and you're done. Yeah, so we'll see.
10:07
I just thought it was an interesting historical
10:09
corollary. And you know, now that you talk
10:11
about it, we should probably round about mid-october come out
10:16
with a episode on stock market crash, given that the two
10:20
crashes we've had of that magnitude have both been in
10:23
October. And I October's the month.
10:26
Yeah, and I seem to recall there's some reason for that.
10:29
It's like there's a whole bunch of important Fed and government
10:33
reports that tend to come out in October.
10:37
Yeah, well, we'll, we'll talk about how reports impact the
10:41
markets in just a minute. But first, first at the
10:45
Foolishness of the Week. Get words and numbers on
10:49
civil.com and on Civil's TV and mobile apps.
10:53
Regular episodes are free. Sign up for an annual membership
10:57
to support our show and for access to member exclusive
11:01
content. I want to draw our attention,
11:04
Anthony, to the WNBA, something that we, we tend not to talk
11:08
about much. And you know, we could talk
11:11
about their crackpot salary demands and, and this sort of
11:14
thing. Pay us what we're worth.
11:17
The, the league has never turned a profit.
11:20
So the players are literally worth less than nothing.
11:23
So they're they're not being paid what they're worth and
11:25
they're quite right about that. From a certain perspective,
11:28
they're, they're getting way more than they're worth.
11:30
But I, I think that's kind of unfair.
11:32
And that's a bit of a slap. These, these are talented women
11:36
who do a thing that is becoming much more desirable to the US
11:43
sporting public. People have started to pay
11:46
attention. And it's not really a surprise
11:49
because people pay attention to the women's NCAA championship
11:53
every year, right? So it's not like this is coming
11:56
out of left field, that one day nobody cared about it and the
11:59
next day, Oh my God, there's a professional league dedicated to
12:03
it. But as with anything that goes
12:09
from not terribly popular to pretty popular, and presumably
12:13
on its way to very popular, there are always some growing
12:16
pains. And people are behaving in the
12:20
worst way at WNBA games. They are throwing sex toys onto
12:27
the floor. Oh my God.
12:28
And that, my friend, should be beyond the pale, no matter what
12:33
you happen to think of women's basketball.
12:37
And you've already paid for your ticket.
12:40
It was it worth it to be a complete and total jerk about
12:44
everything? Did did the 4050 sixty bucks you
12:47
paid to get in the door? Was it worth it?
12:50
Just stay home for crying out loud.
12:53
Now, I'm a little testy about this.
12:55
I am from the great state of Connecticut where it is mandated
12:58
by law that we are all female women's basketball fans.
13:03
So, you know, maybe I take this a little harder than most.
13:05
It's not funny and it should stop, be it.
13:10
So this gets us it to where we should have been.
13:13
Something else that isn't funny and should stop.
13:15
Yeah, something else that's not funny.
13:17
We're we're back to data. We, we got into this just a
13:20
little bit. I guess it was last week, the
13:22
week before, can't remember when the and we have to stop talking
13:26
about last week and the week before because we are now a
13:30
twice a week podcast. So our our vernacular is going
13:34
to have to change a bit. But an episode or two ago, we
13:38
got into the downward revision in the jobs numbers that had
13:43
just freshly come out that day. And people were boy, were they
13:48
pissed. Boy, were they pissed.
13:50
And and Donald Trump was most pissed of all, so much so that
13:55
he fired the woman who was in charge of the Bureau of Labor
13:58
Statistics. Yeah, Yeah, that's a if.
14:02
Yes. In fact, I thought for sure that
14:05
you would, Peg. That is the foolishness of the
14:08
week, and I have a lot to say about it.
14:10
Yeah, I figured it would come up here.
14:12
But here's here's the thing I think people don't realize
14:16
economic numbers are constantly being revised.
14:21
This is not news. It's not like all of a sudden
14:24
BLS, which is Bureau of Labor Statistics.
14:27
It's not that all of a sudden BLS out of the blue decide to
14:30
revise some numbers. This In fact, there's a regular
14:32
schedule for the revisions. And, and let let's stop before
14:37
we go any further and just point out that I, I, I don't care what
14:43
your political persuasion is. Left, right, center, the BLS is
14:48
the gold standard for statistics in the United States.
14:53
Yeah, it's not even close. They're they're leagues ahead of
14:58
anybody else. They.
15:00
They do a good job they. They do a fabulous job in in the
15:04
fact of the matter is James, it's more than just you know, in
15:08
this case it was employment jobs numbers that they were revising.
15:12
It's not just jobs numbers, they're they're measuring all
15:15
sorts of things across the economy.
15:17
And these numbers are used in businesses will use the markets
15:22
use them when you people are pricing stock options or making
15:27
plans to invest in the future. What are they doing?
15:29
They're looking at BLS numbers to try and get a handle of
15:32
what's going on. So this is, you know, to mess
15:35
with BLS numbers is, is a big deal.
15:38
This is a it's, you know, it's like reaching into your car
15:42
engine and, you know, unscrewing a couple turns, the spark plug,
15:46
all kinds of things downstream are going to start breaking
15:49
because of this. And my concern, and I think the
15:52
concern of every rational person is that, OK, the president has
15:57
decided that the woman in charge of BLS has to go because, as he
16:02
said, how can anybody be so wrong?
16:05
Well, all right, take take a giant deep breath.
16:09
Because the real concern is that he puts a yes man into that job
16:13
who tells him what he wants to hear.
16:16
Oh, yeah. Oh, that's horrible.
16:18
It's. And that would absolutely
16:20
destroy the legitimacy of the BLS.
16:23
And I've heard you say, you know as an anarchist in your heart,
16:28
that if you had to get rid of every government agency, BLS
16:32
might be the last one to go. Right.
16:34
Yeah. The guys who are collecting the
16:36
data, those are the ones I'd keep, yeah.
16:38
I I remember you saying this and it stuck with me.
16:41
But you you touch on a serious problem, which is going forward,
16:47
BLS is accuracy is now going to be called into question.
16:52
Because if the president can fire the commissioner of the BLS
16:58
because he doesn't like their numbers, all of a sudden it's
17:02
there's a chilling effect all the way down where everybody is
17:06
going to think twice before reporting a bad number, but
17:09
they're not going to think twice about reporting a good number.
17:12
So all of a sudden the numbers are all inflated in in One
17:15
Direction. You know what?
17:17
This sounds like it. It sounds like every
17:19
authoritarian government that ever existed.
17:22
Oh yeah? Well, this is a big problem that
17:24
we have with China. China's numbers are heavily
17:28
influenced, right? And so it's, it's really
17:30
difficult to get to get a bead on what's actually going on with
17:34
China's economy. Now imagine if we're 25% or more
17:39
of the world's economy. What happens when the world
17:41
can't trust the numbers that we're claiming are true about
17:45
ourselves? Yeah.
17:46
And, and look, this happens in North Korea.
17:48
This happened in the Soviet Union.
17:50
You take your pick. Every authoritarian regime has
17:53
this problem. Why?
17:55
Because if you report something to Illinois duche, whoever
17:59
Illinois Duche happens to be in your specific situation, you
18:03
might get killed. So you're going to choose what
18:05
to report very, very carefully. Yeah.
18:08
And if you. Know no that means.
18:09
You're being killed here. No, but losing your livelihood
18:13
is a. Problem right?
18:15
And it's not just the head of the department, as you say, it's
18:20
a chilling effect all the way down.
18:22
Yeah, yeah. Everybody all the way down is
18:24
going to be reacting to this. So having said all of this as
18:28
prelude, I I know you've gotten into your beloved spreadsheets
18:33
and done. I do have it right.
18:35
In front of me. That thing that you do to
18:38
irritate all of us in your own special way.
18:41
So, you know, walk us through some examples of data that's
18:47
been revised. Yeah.
18:49
So I, I guess to set the stage here, people have to understand
18:53
that when you're collecting data on a country that generates over
19:00
$30 trillion in GDP, it's comprised composed of what, 150
19:07
million households, 300 million plus population, there's no way,
19:13
absolutely zero way you're going to get numbers exactly right.
19:17
There's always going to be some error involved.
19:20
And So what happens is we count some things and we understand
19:25
that the numbers somewhat wrong. And later on we get some revised
19:28
information, we update the number.
19:30
And this happens a couple of times when you're talking about
19:32
jobs numbers, these numbers are coming.
19:35
Hang on one second if if you would yes to everything you just
19:40
said, but I think we and I think you mentioned this in the
19:44
previous episode, we should expect the errors to appear on
19:48
both sides of the line and to more or less cancel each other
19:52
out over. Time, I would expect that to
19:54
happen you would expect for example, the overestimates to
19:59
balance with the underestimates time now that doesn't happen
20:03
here, at least it hasn't been happening with be a lessons
20:06
about 2010. So the the.
20:08
People out there who are saying things like it's all rigged,
20:12
well, it's not, but this? Yeah, yeah, there's, there's,
20:19
they tend to update. I shouldn't say tend to update.
20:22
Well, they do on average, they update in One Direction, but
20:26
that direction is finishingly small.
20:29
The well, let me back up because I'm getting ahead of myself.
20:31
Here's what's, here's how this works.
20:35
We're talking about jobs numbers.
20:37
BLS bases their numbers on on tax forms that businesses
20:44
submit. And of course, for all sorts of
20:47
reasons, some businesses are late in submitting or they have
20:51
a different cycle or what have you.
20:53
So the initial numbers that BLS gets tends to be from the big
20:57
companies, you know, the ones that have accounting departments
21:00
and they're on the ball about reporting their numbers.
21:03
And so initially, that's what BLS has got.
21:06
We've got, we can see how many people are being employed in
21:09
these big companies and BLS will estimate the rest based on
21:13
what's happened in the past. We know that there's a chunk of
21:16
companies out here that plus or minus report so many employees.
21:20
So we include that. And then next month or the month
21:23
after, when those forms finally come in, we can revise our
21:27
numbers and get a better number. So there's actually 2 sets of
21:31
revisions that go on. We get the initial number.
21:33
A month later, we get a revised number based on new information
21:36
that's come in. A month after that, we get
21:38
another revision based on more information that's come in, that
21:41
has come in. And that's basically there's one
21:43
more revision that happens like a year later.
21:45
But basically, that's it, right? And still that numbers not
21:49
sacrosanct. It's not the end result because
21:51
there's all kinds of of noise in the system.
21:54
But it's pretty good. It's pretty good.
21:57
And I can give you an idea of how good it is this revision
22:00
that we're talking about. And So what happened was we're
22:03
talking about June job numbers. The initial estimate was that in
22:09
June, the economy added 147,000 jobs, which is pretty good.
22:16
That number come July got revised downward from 147,000
22:22
new jobs to 14,000 new jobs. Not good.
22:26
Yeah, people lost their minds. I said, where's the 133,000?
22:30
How can you be off by 133,000 jobs?
22:34
Which when I say it like that, does sound like a lot.
22:37
Except 133,000 jobs is 81 hundredths of 1% of all the
22:46
jobs. Really stinking close.
22:50
It is. But the real problem is when you
22:51
have a bunch of politicians saying we're creating new jobs
22:57
and then they see, they see the first number and then they say,
23:00
see, we created new jobs. And then the revision comes out
23:05
and it's nowhere near the original number.
23:08
Now they can't say we've created new jobs.
23:11
Now they have to say we didn't create any jobs or, or work with
23:16
me here for a second. Those people collecting the data
23:19
are involved in nefarious activities.
23:21
Right, right. And much better.
23:23
Much better to say they're involved in nefarious activities
23:26
then to admit what economists coast to coast have been telling
23:29
you since January, that you can't impose 25% tariffs and
23:35
then be changing them all over the place and not expect the
23:38
economy to slow down. There's going to be a
23:40
reverberation of that behavior and and here's the thing that
23:44
most of the tariffs haven't really even been collected yet.
23:48
No, they they haven't. We're at the very beginning of a
23:50
process. Yeah, I was looking the other
23:53
day. The average effective tariff
23:55
rate right now is about 8%, not the 25%.
24:00
And and the reason it's 8% is a couple of things.
24:03
One, only some countries got to 25%.
24:06
I know China got it right off. In fact, China even got a higher
24:09
tariff on some things. There are some countries that
24:12
have much fewer, much lower tariffs.
24:14
Some don't have tariffs at all. And even when you're done with
24:17
all of that, in anticipation of the tariffs, businesses stocked
24:22
up on inventory. Right, which spiked GDP prior to
24:27
tariffs? Right post.
24:30
Tariff. It went way down.
24:32
Yeah. Now, neither of those is the
24:35
correct answer. We're going to have to wait.
24:37
We're going to have to wait and see what the economy does in
24:40
reaction to the tariffs. Indeed, we have to wait and see
24:43
if the tariffs even hold. Legally and politically, they
24:46
might. Not and there is a way out of
24:49
this for President Trump. I don't know that he would take
24:52
it, but the way out is to is to declare victory now, cut all the
24:57
tariffs back to where they were and say, look, see what I did?
25:01
I imposed the tariffs. We had a spike in tax revenues.
25:04
The economy did well. I proven my point.
25:08
That's not going to happen. Yeah.
25:09
But if he holds to these tariffs for another 3-6 months now,
25:14
we're going to really start to see the the negative impact.
25:17
Yeah. But that will be revisions in in
25:20
in the future. I want to talk about revisions
25:22
in the past. Do you have any great examples?
25:24
Well, I, I, I do. And this revision is of the
25:29
133,000, the change of 133,000 jobs.
25:34
I shouldn't, I can say in the one hand it's not unprecedented,
25:39
but on the other hand it is pretty big.
25:41
Now, I, I said it's 81 hundreds of 1% of the entire labor force.
25:46
So in that sense it's a very small number.
25:48
But in terms of revisions, it's pretty small.
25:51
In fact, I went back to 2008 and this downward revision of
25:57
133,000 is the third largest, not the biggest, but it is the
26:01
third largest revision that we've seen since 2008.
26:06
In terms of jobs or in terms of percentage of jobs?
26:09
No, in terms of raw number. Of jobs numbers.
26:12
Yeah, the largest was in March of 2021, which you can imagine
26:17
with all of the nonsense that was going on with COVID, nobody
26:20
knew what was going on. So you would expect some big
26:24
changes. Oh, and by the way, I've left
26:26
out of this entirely pretty much all of 2020 because the
26:32
revisions there were make everything else here look like
26:36
noise. Yeah, they'd.
26:37
Be yeah, they'd be absolutely gargantuan.
26:39
I, I don't even know what to, to think about that you're even now
26:42
five years later. Yeah, yeah, there's a well,
26:45
actually it's interesting because I've, I've come to, when
26:49
I show historic data, I, I've come to show 2012 to 2019.
26:55
Then I do dot dot .2022. I just take out, you have to
26:59
take out 2020-2021 because otherwise the, your graph is
27:03
going like this and all of a sudden you've got this huge jump
27:06
in here and you can't see anything else.
27:07
Yeah. So any, in any event, yeah, it
27:10
was a large, it was large revision, no question.
27:13
It wasn't unprecedented. In fact, some of the larger
27:16
revisions have been happening post COVID, again, not counting
27:21
COVID, the larger revisions are happening post COVID.
27:24
And I think that's because on the one hand we have the
27:28
reverberations of the the shutdown of of the economy and
27:33
then the startup again in the supply chain problems that go on
27:36
and on. You've got that, you've got the
27:38
inflation of 2022 and now we've got the tariffs of 2025.
27:42
It's been since COVID kind of non-stop weirdness going on in
27:47
the economy. So in that sense, I'm not
27:49
surprised that the the major revisions have happened post
27:53
2020. But prior to that, the next
27:56
largest revision you see prior to COVID was back in September
28:00
of 2008. And of course, we all know what
28:02
was going on then. That was the housing crisis.
28:05
Yep, talk about reverberations throughout the economy.
28:09
Words and Numbers is part of the Civil Network.
28:12
Civil is a streaming platform dedicated to upgrading
28:16
civilization by exploring better ways for people to organize,
28:21
innovate, and thrive. At Civil, we believe that
28:25
societies flourish when their members honor the civic equality
28:29
of all people and engage with one another through cooperation.
28:33
Civil seeks great minds from around the world to share bold
28:37
ideas for fostering healthy, dynamic societies.
28:41
By creating space for honest disagreement among people who
28:44
share a commitment to building a better future, Civil offers a
28:48
refreshing alternative to divisive punditry and insular
28:53
echo chambers. Words and Numbers is freely
28:55
available, but those who subscribe to Civil will access
28:58
exclusive content, including regular live streams and
29:02
conversations with remarkable people who are working to build
29:06
a better society. For more information and to
29:09
subscribe, visit civil.com. That's civl.com.
29:14
Use the code Words and numbers. That's one word for 50% off an
29:19
annual subscription. But I think that the take home
29:23
here is a couple of things. One, yeah, you've got this large
29:28
revision in terms of numbers, but in terms of of fraction of
29:33
the workforce, it's really small in in fact, it's like, it's like
29:38
throwing a dot, a dart at a dart board and you didn't get it
29:42
right in the center, but you're still on that red patch, right?
29:44
You're that close. For all intents and purposes,
29:46
you got the number right. So that's one thing to think
29:48
about. Another thing to think about is
29:50
it matters how we count these things.
29:53
Let's see, let me say it differently.
29:56
It's not entirely clear what the right way to count these things
30:01
is. And so we count them multiple
30:02
ways. And your knee jerk reaction
30:05
might be, well, how hard can it be?
30:07
We're counting jobs. Oh, it could be really, really
30:10
hard. I mean, look, tell, tell the
30:13
people about how unemployment is calculated.
30:17
Oh, yeah, What the problem with unemployment is you've got
30:19
somebody who doesn't have a job. You have to determine why they
30:23
don't have a job. Do they have a job?
30:25
Because. And you can't do that.
30:28
No, that becomes really, really hard.
30:30
In a country of 350 million people, you can't call everybody
30:35
who doesn't have a job and say, hey Joe, why don't you have a
30:38
job? Well, but even if I could, how
30:41
can I be sure that they're telling me the right reason that
30:43
for example. Exactly right.
30:45
If you don't have a job because you're looking as hard as you
30:48
can, but there's just nothing out there for you, OK, you're
30:51
unemployed. That's what if you don't have
30:53
it. That's unemployed, Yeah.
30:54
If you don't have a job because you don't want one, well, that's
30:57
not unemployed. Not unemployed, but you are
31:00
unemployed, right But, and therefore.
31:02
Yeah, you're not right. You're what an economist would
31:06
call non employed. Non employed but you, but
31:08
nonetheless. You in the unemployment numbers.
31:10
We in the general population would say, yeah, that's
31:13
unemployment, but it isn't and but it isn't right.
31:18
And, and there's a number of things that share in this
31:20
nature. I think unemployment is just the
31:22
easiest one to. Illustrate what it is.
31:24
And let me tell you the problem here, because again, you might
31:28
think it's very easy. We'll just count the number of
31:30
people who are working if you want to know the number of jobs.
31:32
Yeah, but how do you want to do that?
31:34
Do you want to ask the employers how many people are you
31:37
employing? You can do that.
31:40
But notice a problem if you've got one guy who holds down to
31:44
jobs, he shows up twice. And so it looks like you've got
31:49
two people working here when in fact you only have one guy, but
31:52
he's holding down to jobs. So when we ask employers, we're
31:55
counting jobs. That's the number that got
31:57
revised. When you another way you can do
32:00
it is you call households and you can ask do you have or are
32:04
you employed? And the problem there is if the
32:08
guy has again two jobs, he's going to say, yes, I'm employed.
32:14
So you're counting one when in fact it's 2.
32:17
So you you do it the one way and you get a 2 when it should be 1.
32:21
You do the other way, you get a one way, it should be two.
32:23
And and neither number is entirely right nor is it
32:27
entirely wrong. They're trying to get it.
32:30
This I'll defined thing from two different perspectives.
32:33
And data being what it is, spread it it It's generally just
32:37
binary yes or no. Right.
32:40
Yeah, yeah. Do you have a job, yes or no?
32:43
And and now we've got a problem. Yeah, well, yeah, this isn't
32:47
this. Isn't, generally speaking, a
32:49
huge problem? People who work with data
32:52
understand all of this. Oh yeah.
32:54
It's it's no big deal. And the reason why we count
32:58
things multiple ways is to get a better vision of something from
33:01
another angle, right? So it makes perfect sense to
33:05
have multiple ways to count certain things that at first
33:09
blush may be intuitively obvious, right, but on 2nd
33:14
consideration become very, very complicated.
33:18
Yeah, and the numbers are going to be in the same ballpark.
33:21
It's not like they're going to get 100 million off of 1 number
33:24
and 10,000 off the other, right? They're going to be roughly the
33:27
unemployment rate is never 3% one way and 87% the other,
33:33
right? Yeah.
33:34
I mean that that's just ridiculous, right?
33:36
That would never happen. Words and Numbers is a Civil
33:40
original podcast. We work with Civil to advance
33:43
their mission of upgrading civilization by exploring ideas
33:48
that promote civic equality and cooperation.
33:51
We believe the most pressing issues humans face, war,
33:55
poverty, debt, tribalism, arise from coercive institutions that
34:00
bring out the worst in us. We at Civil look to voluntary
34:04
solutions to these and other problems.
34:07
We invite you to explore Civil's new platform at civil.com.
34:12
That's civl.com. Yeah.
34:17
So really this is about expectations people have.
34:22
And you know, there's this old saying people who don't work in
34:26
statistics say, well, there are are lies, damn lies and
34:29
statistics. Well, that that's not that's not
34:32
exactly fair. No, it's it's not because the
34:36
fact the matter is. I get the point, but it's not
34:40
exactly fair. Well, it it's not because you
34:43
know, people will say things like, well, you can get
34:45
statistics to say anything you want them to say, and that's not
34:48
correct. I can, I can lie to you, I can
34:53
misrepresent the statistics, but I can't make them say what I
34:56
want them to say unless you do something like President Trump
35:01
just did. Well, fire the guy who's trying
35:04
to revise the statistics. I I, I don't want to get too far
35:07
into the statistical weeds, but people who say that are on to
35:11
something. They probably had a a college
35:15
level insurance stats course and learned that if you change the
35:19
confidence interval you get different.
35:21
Results. Sure.
35:22
Yeah, but, but, but again, if I'm a, if I'm a a reasonable
35:28
statistician, I'm going to explain to you what just
35:31
happened here so that you can understand.
35:33
It is not that the reality changed, it's that you're now
35:36
looking at it from a different viewpoint.
35:38
But if I take the confidence level and I, I, I change it
35:42
radically, you know, to, oh, say 75%, which which no statistician
35:47
would ever do, you're, you're going to get numbers that make
35:50
absolutely no sense when you look.
35:52
At reality, right? So you could have dishonest
35:55
statisticians, yes, but you don't have dishonest statistics.
35:59
And and the standards are pretty well prescribed at this point.
36:04
They are in, and you raise a, a point that's worth mentioning.
36:08
There's tremendous transparency at the BLS Bureau of Labor
36:12
Statistics. Not only is there transparency
36:14
that in fact, you, I, I can go, I got these numbers off their
36:17
website. I went to their website and said
36:19
specifically show me all of your data revisions going back as far
36:23
as you can go. And you know, here it is.
36:25
It's not a big deal. You can get this stuff.
36:27
Not only that, but they have hundreds of economists working
36:31
on this, data economist, data analysts.
36:34
So it's not like you've got, you know, two or three people in a
36:37
backroom somewhere who can change these numbers from that
36:41
way. Well yeah, the only way you can
36:43
mess with it is to make everybody afraid of their jobs
36:47
for reporting numbers that the higher ups won't like.
36:50
Yeah, that's right, you will have to get statisticians to
36:55
lie. And remember, you know, now that
36:57
we're having this conversation, I'm, I'm thinking to myself
36:59
this, it seems like I've had this conversation before.
37:03
And I did back in 2020 when CDC was releasing data on, on COVID
37:13
infections and deaths and so forth.
37:15
And data analysts, we're looking at that at the time saying, I'm
37:19
not sure that there's something right here.
37:22
And we're, you know, you, you argue about how the data was
37:25
collected and it's all above board.
37:26
And that's fine until someone comes along and says, you're
37:30
saying things I don't want you to say.
37:32
So you're going to lose your job.
37:33
Then that just comes a. Problem.
37:34
You have just opened a can of worms.
37:36
We will get hate mail. Send it to Anthony.
37:41
But here's the thing, I, I think the transparency is incredibly
37:45
important. It's incredibly important that
37:46
you leave these people alone, let them do their jobs free of,
37:51
of pressure. And that's how you get the,
37:53
you're not going to get perfect numbers, but it's how you get
37:56
the best numbers on an ongoing basis.
37:59
Yeah, I think that's right. You got any other numbers you
38:01
want to throw? At us, not at the moment.
38:03
This is the big, big thing I was looking at.
38:06
And you know, again there the revision was 8 tenths of 1% of
38:11
total jobs. So I think the So what here is
38:17
so obvious as to be inescapable. Well, you need good statistics
38:25
if you're going to inform policy.
38:29
We're going to do X because A and we'll do Y if B.
38:35
But if you don't get A&B right, X&Y are going to be a disaster.
38:40
Yeah, and the the trick, I think the key honestly is
38:42
transparency. You'd just be transparent.
38:44
About it, it has to be whether you get numbers you like or
38:48
numbers you don't. That's not the question, right?
38:51
The question is, what do we do about it, right?
38:54
And, and sometimes if you say become become president in
38:59
January and by the time you get to August, that's really not a
39:04
long time. Politically, it seems like a
39:07
long time in this administration where things just happen minute
39:12
by minute. It seems almost.
39:14
But do you really expect an incoming president to be able to
39:18
influence jobs in six months? You and I are on record as
39:23
saying presidents don't influence jobs that much at all
39:26
anyway. Right.
39:27
Yeah. Right and at.
39:30
The poll, he's start doing things like this president's
39:32
doing. In the negative direction they
39:35
do, in the positive direction, not as much.
39:38
Yeah, yeah. And I think that's pretty fair.
39:43
But in order to do anything well, you have to start with,
39:49
dare I say it, the truth. Even if you don't like it, there
39:55
are all kinds of truths in this world that I don't like at all.
39:59
I have to take account of them if I'm to do anything well.
40:02
And that's, that's where we're at.
40:04
And it's a little heartbreaking to me that everybody doesn't
40:08
just see it that way as a matter of course.
40:10
Well, I'm interested in what happens next.
40:12
I don't know. Has he named a successor to
40:16
Erica Mcconterfer I believe? Not to not to my knowledge.
40:23
Yeah. So the question is, what happens
40:25
with BLS moving forward if he appoints someone who is
40:29
obviously a kind of loyal to him?
40:35
Maybe a Fox News host like he did with the Department of
40:39
Defense, right? That would raise my eyes a
40:43
little bit. It's a little bit disturbing
40:45
because we're seeing this here. We're also seeing it with the
40:47
Fed, his calls to fire the the Fed chair.
40:53
And I believe we just had one of the, one of the members of the
40:56
Fed, I believe just resigned couple of days ago.
41:00
Yeah, we, we have to be very careful here.
41:04
Those institutions not, not only do they need to be independent,
41:10
maybe more importantly, they have to be perceived as
41:14
independent by people on both sides of the aisle.
41:17
Yeah, and they have to be consistent over time and.
41:20
I'll, I'll remind people who, who perhaps are cheering that,
41:25
you know, the president is doing all these things.
41:28
I'll remind them be, be very afraid.
41:31
Because if we allow a president to get away with this sort of
41:36
thing, putting pressure on the Fed, putting pressure on the
41:39
Bureau of Labor Statistics, the next guy who's elected, who
41:44
maybe you wish hadn't been elected, is going to have those
41:48
same powers. He will.
41:49
He will and he will also or she will also do the exact same
41:53
nonsense. Yeah, just in the other
41:55
direction. Because that's the pernicious
41:57
nature of power. It's not about this or that.
42:01
It's always about power. And the the thing that the
42:04
American founders, the framers were so good at was seeing that
42:08
problem clearly. Yes.
42:11
And we don't see it at all anymore.
42:14
We don't think power is the problem.
42:16
We think power in the hands of our opponents is the problem.
42:20
Yeah, yeah. But the real problem is power.
42:22
Anyway, on that happy note, I will remind you that whether you
42:26
agree with us or disagree with us, that's not the issue today.
42:29
The issue today is how nice are you going to be to somebody else
42:34
out there? And I'll just say, take positive
42:36
steps. Go try to do something nice that
42:39
you wouldn't have otherwise done.
42:41
And for the rest of you who can't get up to that standard,
42:44
just be nice to each other. We'll see you next week.
42:47
See you next weekend. See you next week, James.