WEBVTT

00:00:16.879 --> 00:00:22.199
So, today I'm going to talk about the

00:00:18.320 --> 00:00:24.399
Phillips curve and inflation. Um

00:00:22.199 --> 00:00:25.320
Now, as I said in the previous lecture,

00:00:24.399 --> 00:00:27.159
uh

00:00:25.320 --> 00:00:30.039
the material that is specific to this

00:00:27.160 --> 00:00:31.920
lecture will not enter this quiz.

00:00:30.039 --> 00:00:33.359
It's the beginning of what is perhaps

00:00:31.920 --> 00:00:37.160
the most important model you'll see in

00:00:33.359 --> 00:00:38.920
the in in this in this uh class, but

00:00:37.159 --> 00:00:42.039
it will take us uh three or four

00:00:38.920 --> 00:00:44.600
lectures to to develop. So, I'm going to

00:00:42.039 --> 00:00:45.320
say things that certainly will

00:00:44.600 --> 00:00:46.920
um

00:00:45.320 --> 00:00:49.679
may help you understand a little better

00:00:46.920 --> 00:00:51.560
the previous lecture, and so

00:00:49.679 --> 00:00:52.799
if you're only concerned about the next

00:00:51.560 --> 00:00:55.240
quiz,

00:00:52.799 --> 00:00:57.679
uh there will be a sort of uh small

00:00:55.240 --> 00:01:00.120
review of the previous lecture here. Uh

00:00:57.679 --> 00:01:02.359
but again, anything that's specific to

00:01:00.119 --> 00:01:03.280
this lecture and was not in the previous

00:01:02.359 --> 00:01:06.719
one

00:01:03.280 --> 00:01:08.280
won't be part of of this quiz.

00:01:06.719 --> 00:01:10.239
So, what is this Phillips curve? Well,

00:01:08.280 --> 00:01:14.320
in in uh

00:01:10.239 --> 00:01:16.239
in 1958, an an economist uh at LSE, the

00:01:14.319 --> 00:01:17.879
London School of Economics,

00:01:16.239 --> 00:01:20.439
came up with some just an empirical

00:01:17.879 --> 00:01:24.640
relationship. This is A.W. Phillips. He

00:01:20.439 --> 00:01:28.319
found that using historical data

00:01:24.640 --> 00:01:30.760
uh for the US, I think he did it. Um

00:01:28.319 --> 00:01:33.639
uh there was a negative relation up to

00:01:30.760 --> 00:01:35.160
sort of the '50s, I think. Uh

00:01:33.640 --> 00:01:37.799
the there was a negative relation

00:01:35.159 --> 00:01:39.759
between uh the unemployment rate and the

00:01:37.799 --> 00:01:42.479
rate of inflation.

00:01:39.760 --> 00:01:44.040
And then our very own Paul Samuelson and

00:01:42.480 --> 00:01:46.240
Robert Solow

00:01:44.040 --> 00:01:48.280
labeled this relationship the Phillips

00:01:46.239 --> 00:01:49.759
curve in honor of uh

00:01:48.280 --> 00:01:52.000
A.W. Phillips.

00:01:49.760 --> 00:01:55.880
And nowadays it's sort of is a central

00:01:52.000 --> 00:01:58.040
concept uh in macroeconomics, and

00:01:55.879 --> 00:02:00.399
uh and uh it's certainly very, very

00:01:58.040 --> 00:02:02.719
relevant to understand what is going on

00:02:00.400 --> 00:02:04.480
uh right now in not only in the US

00:02:02.719 --> 00:02:06.159
economy, but in most economies around

00:02:04.480 --> 00:02:07.280
the world.

00:02:06.159 --> 00:02:08.960
So, let me

00:02:07.280 --> 00:02:10.520
show you sort of this is not the one

00:02:08.960 --> 00:02:12.480
that uh

00:02:10.520 --> 00:02:14.960
that Phillips uh plotted. I think this

00:02:12.479 --> 00:02:15.519
is the one that uh

00:02:14.960 --> 00:02:17.920
uh

00:02:15.520 --> 00:02:20.960
Samuelson and Solow plotted for data

00:02:17.919 --> 00:02:23.879
from between 1900 and 1960 uh

00:02:20.960 --> 00:02:25.920
for the US, you found you find sort of

00:02:23.879 --> 00:02:27.519
this sort of negative correlation. I

00:02:25.919 --> 00:02:28.479
think it's reasonable.

00:02:27.520 --> 00:02:29.480
Uh

00:02:28.479 --> 00:02:31.679
um

00:02:29.479 --> 00:02:34.519
there's negative correlation between

00:02:31.680 --> 00:02:36.120
uh the unemployment rate and inflation

00:02:34.520 --> 00:02:37.760
rate, no? At very low levels of

00:02:36.120 --> 00:02:39.759
unemployment, you typically see very

00:02:37.759 --> 00:02:41.759
high levels of inflation.

00:02:39.759 --> 00:02:44.000
Conversely, sort of at very high levels

00:02:41.759 --> 00:02:46.719
of unemployment, you tend to receive low

00:02:44.000 --> 00:02:49.000
levels of inflation or even deflation.

00:02:46.719 --> 00:02:52.159
In fact, this period includes the the

00:02:49.000 --> 00:02:53.960
Great Depression, for example.

00:02:52.159 --> 00:02:56.400
So,

00:02:53.960 --> 00:02:59.040
that's sort of the data. And and again,

00:02:56.400 --> 00:03:00.960
this was just an empirical regularity.

00:02:59.039 --> 00:03:03.840
But we can build some theory about this

00:03:00.960 --> 00:03:05.560
relationship using the ingredients most

00:03:03.840 --> 00:03:07.319
of the ingredients that

00:03:05.560 --> 00:03:10.599
I mean, essentially we can build a

00:03:07.319 --> 00:03:13.039
relationship that is downward sloping

00:03:10.599 --> 00:03:14.239
from the ingredients we already have.

00:03:13.039 --> 00:03:15.679
And this is the part that is a little

00:03:14.240 --> 00:03:17.280
bit of a review

00:03:15.680 --> 00:03:20.879
of the previous lecture.

00:03:17.280 --> 00:03:22.759
Remember that we had um um

00:03:20.879 --> 00:03:25.199
actually the previous two lectures. We

00:03:22.759 --> 00:03:28.239
had a wage setting equation

00:03:25.199 --> 00:03:29.639
W equal expected prices

00:03:28.240 --> 00:03:31.400
and then a decreasing function of

00:03:29.639 --> 00:03:34.279
unemployment and an increasing function

00:03:31.400 --> 00:03:36.080
of uh these labor market supporting

00:03:34.280 --> 00:03:38.479
institutions or

00:03:36.080 --> 00:03:40.840
worker supporting institutions, that

00:03:38.479 --> 00:03:43.119
institutional variables, I should say.

00:03:40.840 --> 00:03:46.439
And um and then we had a price setting

00:03:43.120 --> 00:03:48.000
equation, which was simply the wage uh

00:03:46.439 --> 00:03:50.960
marked up.

00:03:48.000 --> 00:03:53.120
M is a positive constant. So, let me

00:03:50.960 --> 00:03:55.920
start from these two what So, what I'm

00:03:53.120 --> 00:03:57.400
trying to do is derive a Phillips curve.

00:03:55.919 --> 00:03:59.559
Again, this was only an empirical

00:03:57.400 --> 00:04:01.319
relationship, but it turns out that even

00:03:59.560 --> 00:04:03.920
with theory we knew by the time of, you

00:04:01.319 --> 00:04:05.479
know, Samuelson and and and Solow, we

00:04:03.919 --> 00:04:07.679
could sort of come up with a with a

00:04:05.479 --> 00:04:09.359
theory of that relationship. And that

00:04:07.680 --> 00:04:11.280
theory builds on the ingredients we have

00:04:09.360 --> 00:04:12.960
been looking at. So, these are the price

00:04:11.280 --> 00:04:14.719
set the wage setting equation, the price

00:04:12.960 --> 00:04:16.920
setting equation. I'm going to just

00:04:14.719 --> 00:04:19.519
simplify things and assume that this

00:04:16.920 --> 00:04:22.079
this relationship here, this function f

00:04:19.519 --> 00:04:24.680
of u z, is some linear function, at

00:04:22.079 --> 00:04:26.519
least locally linear function, uh which

00:04:24.680 --> 00:04:28.720
is decreasing in unemployment and

00:04:26.519 --> 00:04:32.359
increasing in z.

00:04:28.720 --> 00:04:32.360
Why is it decreasing in unemployment?

00:04:35.319 --> 00:04:39.040
This says, no, that that if unemployment

00:04:37.759 --> 00:04:41.639
goes up,

00:04:39.040 --> 00:04:43.640
for any given expected price,

00:04:41.639 --> 00:04:46.479
wage demand is lower.

00:04:43.639 --> 00:04:49.560
Okay? And that's essentially because uh

00:04:46.480 --> 00:04:50.840
for the worker is is sort of is a

00:04:49.560 --> 00:04:52.280
becoming unemployed is a scary

00:04:50.839 --> 00:04:54.199
situation.

00:04:52.279 --> 00:04:56.279
Conversely, for firms it's higher it's

00:04:54.199 --> 00:04:56.839
easier to find uh

00:04:56.279 --> 00:04:59.279
uh

00:04:56.839 --> 00:05:01.279
a worker and and uh

00:04:59.279 --> 00:05:04.279
and uh So, as we said, a worker is

00:05:01.279 --> 00:05:05.679
scared for two reasons. One is that it's

00:05:04.279 --> 00:05:07.879
more likely it gets fired when

00:05:05.680 --> 00:05:10.879
unemployment is high, typically that's a

00:05:07.879 --> 00:05:13.639
recession. It's also likely they know

00:05:10.879 --> 00:05:15.759
that that worker knows that if she were

00:05:13.639 --> 00:05:17.319
to fall into the unemployment pool, it

00:05:15.759 --> 00:05:19.759
would take a longer time to get out of

00:05:17.319 --> 00:05:21.639
it. Okay? And the firms are seeing the

00:05:19.759 --> 00:05:24.000
opposite side. It's pretty easy for them

00:05:21.639 --> 00:05:26.279
to replace a worker if they were to

00:05:24.000 --> 00:05:28.839
dismiss a worker because there's lots of

00:05:26.279 --> 00:05:30.759
available workers in unemployment. Okay?

00:05:28.839 --> 00:05:32.279
So, that's the reason that's negative.

00:05:30.759 --> 00:05:34.680
So, I'm going to stick this function

00:05:32.279 --> 00:05:37.519
back in here.

00:05:34.680 --> 00:05:39.720
And then I'm going to replace this W

00:05:37.519 --> 00:05:42.079
with this function in there in the price

00:05:39.720 --> 00:05:46.800
setting equation, and I end up with an

00:05:42.079 --> 00:05:50.399
equation for P. Okay? So, this says that

00:05:46.800 --> 00:05:52.480
the price, given the expected price,

00:05:50.399 --> 00:05:54.239
is decreasing in unemployment and

00:05:52.480 --> 00:05:55.920
increasing in z and increasing in the

00:05:54.240 --> 00:05:58.879
markup.

00:05:55.920 --> 00:06:01.840
So, again, why is this price decreasing

00:05:58.879 --> 00:06:01.839
in unemployment?

00:06:02.600 --> 00:06:05.800
This is the part that is review of the

00:06:04.079 --> 00:06:07.319
previous lecture.

00:06:05.800 --> 00:06:09.319
Previous today. Because

00:06:07.319 --> 00:06:10.879
the wages go down and then the labor uh

00:06:09.319 --> 00:06:14.159
factors of production are cheaper. Okay,

00:06:10.879 --> 00:06:16.279
perfect. Because wages go down, since uh

00:06:14.160 --> 00:06:17.520
our firm needs one worker to produce one

00:06:16.279 --> 00:06:19.759
unit of a good, then the cost of

00:06:17.519 --> 00:06:22.759
production of one unit goes down with

00:06:19.759 --> 00:06:24.399
the wage, and and therefore the price

00:06:22.759 --> 00:06:26.879
goes down. Because the firm is asking

00:06:24.399 --> 00:06:31.359
for a constant markup over that wage,

00:06:26.879 --> 00:06:35.120
the wage declines, and the price drops.

00:06:31.360 --> 00:06:37.400
Good. So, that's all review. So,

00:06:35.120 --> 00:06:39.959
this equation you had seen just without

00:06:37.399 --> 00:06:42.079
an explicit functional form here. What I

00:06:39.959 --> 00:06:43.479
want to do is to go from here. This is

00:06:42.079 --> 00:06:44.839
still not the Phillips curve. Remember,

00:06:43.480 --> 00:06:46.759
the Phillips curve was a relationship

00:06:44.839 --> 00:06:48.519
between inflation

00:06:46.759 --> 00:06:51.279
and unemployment. Here we have a

00:06:48.519 --> 00:06:53.359
relationship between the price level

00:06:51.279 --> 00:06:55.639
and unemployment. Okay? So, we want to

00:06:53.360 --> 00:06:57.560
take one

00:06:55.639 --> 00:06:58.680
one derivative higher. We want to go to

00:06:57.560 --> 00:07:00.079
relation between

00:06:58.680 --> 00:07:03.759
inflation and unemployment. And

00:07:00.079 --> 00:07:04.560
inflation is is the rate of change of P.

00:07:03.759 --> 00:07:07.000
No?

00:07:04.560 --> 00:07:09.519
It's not it's not a level of P.

00:07:07.000 --> 00:07:11.639
So, to do that, all that we'll do is So,

00:07:09.519 --> 00:07:12.680
this when I don't have a subscript here,

00:07:11.639 --> 00:07:14.599
I mean

00:07:12.680 --> 00:07:16.639
the price at time t.

00:07:14.600 --> 00:07:18.360
Okay? And this is the expected price for

00:07:16.639 --> 00:07:18.959
next period, that's what we have.

00:07:18.360 --> 00:07:21.319
Uh

00:07:18.959 --> 00:07:22.239
but today for next period.

00:07:21.319 --> 00:07:25.120
What I'm going to do, I'm going to

00:07:22.240 --> 00:07:27.840
divide both sides by P minus 1. By that,

00:07:25.120 --> 00:07:29.360
I mean the price in the previous period.

00:07:27.839 --> 00:07:32.199
Okay? So, both sides. I'm going to

00:07:29.360 --> 00:07:34.759
divide this side by P minus 1.

00:07:32.199 --> 00:07:36.920
And and and this one by P minus 1.

00:07:34.759 --> 00:07:39.759
So, I get that expression. Okay? That's

00:07:36.920 --> 00:07:41.520
exactly the same equation we had before.

00:07:39.759 --> 00:07:44.560
All that I did is I divided by P minus

00:07:41.519 --> 00:07:46.120
1. Remember remember what this means.

00:07:44.560 --> 00:07:49.120
So, if this is the price for the

00:07:46.120 --> 00:07:50.079
beginning for January 2023,

00:07:49.120 --> 00:07:51.879
uh

00:07:50.079 --> 00:07:54.919
this is the price for, say, where we're

00:07:51.879 --> 00:07:56.719
using annual data, for January 2022.

00:07:54.920 --> 00:08:00.240
Okay? So, I'm dividing by the price of

00:07:56.720 --> 00:08:03.520
January 2022 both sides.

00:08:00.240 --> 00:08:06.800
Now, notice that remember that I can

00:08:03.519 --> 00:08:08.639
that P over P minus 1 is equal to 1 plus

00:08:06.800 --> 00:08:10.759
the inflation rate. Remember where

00:08:08.639 --> 00:08:14.279
inflation rate is just

00:08:10.759 --> 00:08:16.279
P minus P minus 1 over P minus 1. So,

00:08:14.279 --> 00:08:17.519
this is just straightforward algebra,

00:08:16.279 --> 00:08:21.279
no?

00:08:17.519 --> 00:08:21.279
Remember our definition of inflation.

00:08:33.799 --> 00:08:39.519
That's P minus that's inflation. Okay?

00:08:36.440 --> 00:08:41.039
So, 1 plus pi is just

00:08:39.519 --> 00:08:43.079
P minus uh

00:08:41.038 --> 00:08:44.319
over P minus 1. Okay? And that's what

00:08:43.080 --> 00:08:45.720
you have there.

00:08:44.320 --> 00:08:48.400
I can do the same for expected

00:08:45.720 --> 00:08:51.279
inflation.

00:08:48.399 --> 00:08:51.279
Notice that

00:08:53.399 --> 00:08:58.279
sometimes people get confused, but

00:08:56.120 --> 00:09:00.320
expected inflation

00:08:58.279 --> 00:09:02.199
is equal to P

00:09:00.320 --> 00:09:05.120
expected

00:09:02.200 --> 00:09:07.640
not minus P expected minus 1. It's P

00:09:05.120 --> 00:09:10.200
minus 1.

00:09:07.639 --> 00:09:12.080
P minus 1.

00:09:10.200 --> 00:09:15.200
And the reason I'm not subtracting the

00:09:12.080 --> 00:09:16.240
expectation here is because at time t,

00:09:15.200 --> 00:09:18.200
which is when you're forming that

00:09:16.240 --> 00:09:21.000
expectation, you already know what

00:09:18.200 --> 00:09:22.680
happened at t minus 1.

00:09:21.000 --> 00:09:25.440
Okay? So, that's the reason this is

00:09:22.679 --> 00:09:26.559
expected inflation. I don't I don't need

00:09:25.440 --> 00:09:28.480
uh uh

00:09:26.559 --> 00:09:31.000
um

00:09:28.480 --> 00:09:33.080
to put expectations in here. Okay? So,

00:09:31.000 --> 00:09:35.480
that's pi e.

00:09:33.080 --> 00:09:37.520
And so, what we get is

00:09:35.480 --> 00:09:40.279
uh I can replace

00:09:37.519 --> 00:09:42.759
this guy here for 1 plus pi, this guy

00:09:40.279 --> 00:09:45.919
here for 1 plus pi e,

00:09:42.759 --> 00:09:47.919
and I get the following relationship.

00:09:45.919 --> 00:09:49.679
Okay?

00:09:47.919 --> 00:09:52.240
All that I've done is substituting this

00:09:49.679 --> 00:09:54.279
for that, that for that.

00:09:52.240 --> 00:09:58.200
So, that's our price setting equation

00:09:54.279 --> 00:10:00.720
now expressed in terms of inflation rate

00:09:58.200 --> 00:10:02.680
unexpected inflation rate

00:10:00.720 --> 00:10:04.320
and now you know

00:10:02.679 --> 00:10:05.919
if not we're not in Argentina we're in

00:10:04.320 --> 00:10:07.720
the US

00:10:05.919 --> 00:10:09.199
inflation if expected inflation is a

00:10:07.720 --> 00:10:11.840
small numbers

00:10:09.200 --> 00:10:15.120
and the log of 1 plus a small number is

00:10:11.840 --> 00:10:15.120
approximately that number

00:10:15.960 --> 00:10:19.120
so

00:10:17.279 --> 00:10:23.439
I'm going to use this approximation

00:10:19.120 --> 00:10:26.360
which again is valid for X small

00:10:23.440 --> 00:10:30.440
and and so I can replace this 1 plus pi

00:10:26.360 --> 00:10:33.519
for pi and this 1 plus pi e for pi e

00:10:30.440 --> 00:10:36.440
this 1 plus m for m

00:10:33.519 --> 00:10:39.879
plus no and this term here if these

00:10:36.440 --> 00:10:41.760
numbers are not too large again plus

00:10:39.879 --> 00:10:43.399
minus alpha u

00:10:41.759 --> 00:10:45.840
plus z

00:10:43.399 --> 00:10:48.679
and that I do all that and I end up with

00:10:45.840 --> 00:10:48.680
this expression

00:10:51.120 --> 00:10:56.039
all that I've done is I took logs of

00:10:53.120 --> 00:10:59.039
this so I get log of 1 plus pi equal to

00:10:56.039 --> 00:11:02.719
log of 1 plus pi e plus log of 1 plus m

00:10:59.039 --> 00:11:07.399
plus log of 1 minus alpha u plus alpha z

00:11:02.720 --> 00:11:09.399
I'm saying if pi pi e m alpha u plus z

00:11:07.399 --> 00:11:11.559
are not very large numbers which we're

00:11:09.399 --> 00:11:13.600
going to assume then this is

00:11:11.559 --> 00:11:16.159
approximately right so I can rewrite

00:11:13.600 --> 00:11:17.320
that expression as that

00:11:16.159 --> 00:11:20.000
approximately I should have put an

00:11:17.320 --> 00:11:20.000
approximately

00:11:22.000 --> 00:11:25.960
okay so now we have something that looks

00:11:23.679 --> 00:11:27.399
a lot more than like the empirical

00:11:25.960 --> 00:11:29.560
relationship we were talking about we

00:11:27.399 --> 00:11:32.079
have a relationship between inflation

00:11:29.559 --> 00:11:34.039
and unemployment so this says that

00:11:32.080 --> 00:11:35.600
for any given expected inflation and

00:11:34.039 --> 00:11:37.559
markups and and labor market

00:11:35.600 --> 00:11:39.279
institutions

00:11:37.559 --> 00:11:40.559
higher unemployment means lower

00:11:39.279 --> 00:11:43.240
inflation

00:11:40.559 --> 00:11:43.239
why is that

00:11:43.399 --> 00:11:47.319
so that curve tells you that's a

00:11:44.720 --> 00:11:51.920
negative relation we wanted no it says

00:11:47.320 --> 00:11:51.920
higher unemployment lower inflation

00:11:53.600 --> 00:11:56.159
why is that

00:12:02.039 --> 00:12:05.799
look you had it very clear when we talk

00:12:04.159 --> 00:12:07.919
about this no

00:12:05.799 --> 00:12:10.120
you you understood very clearly why an

00:12:07.919 --> 00:12:13.519
increase in unemployment lower the wage

00:12:10.120 --> 00:12:15.720
you understood very clearly why

00:12:13.519 --> 00:12:18.319
therefore an increase in unemployment

00:12:15.720 --> 00:12:20.120
lower the price

00:12:18.320 --> 00:12:22.160
I haven't done anything but algebra in

00:12:20.120 --> 00:12:23.639
the two steps so the same economics

00:12:22.159 --> 00:12:26.600
behind the explanations that you had

00:12:23.639 --> 00:12:29.879
before apply to this curve here

00:12:26.600 --> 00:12:31.759
so the reason inflation will be lower

00:12:29.879 --> 00:12:33.439
when you unemployment is higher given

00:12:31.759 --> 00:12:35.838
all the rest

00:12:33.440 --> 00:12:38.160
is because there's really less wage wage

00:12:35.839 --> 00:12:39.800
pressure workers will demand lower wages

00:12:38.159 --> 00:12:41.639
that means lower prices and therefore

00:12:39.799 --> 00:12:43.319
inflation will be lower the economics

00:12:41.639 --> 00:12:45.240
hasn't changed at all I only I only

00:12:43.320 --> 00:12:47.720
divided both sides

00:12:45.240 --> 00:12:49.759
by p minus 1 and I took the logs and I

00:12:47.720 --> 00:12:50.759
approximated so so the economics has not

00:12:49.759 --> 00:12:53.000
changed

00:12:50.759 --> 00:12:55.799
it just did a little bit of

00:12:53.000 --> 00:12:57.759
basic math okay

00:12:55.799 --> 00:12:59.439
so all the what I'm trying to say is all

00:12:57.759 --> 00:13:02.000
the intuitions that you can already you

00:12:59.440 --> 00:13:03.960
already had from the wage setting price

00:13:02.000 --> 00:13:06.279
setting equations and so on you can

00:13:03.960 --> 00:13:08.560
apply to the Phillips curve as well

00:13:06.279 --> 00:13:08.559
okay

00:13:08.600 --> 00:13:13.800
good so now we have something that

00:13:11.200 --> 00:13:16.080
in principle could explain the type of

00:13:13.799 --> 00:13:17.838
relationship that Phillips

00:13:16.080 --> 00:13:22.800
found and then Samuelson Solo

00:13:17.839 --> 00:13:22.800
corroborated with extended data

00:13:23.279 --> 00:13:25.799
so

00:13:26.200 --> 00:13:30.280
let's let's see how do we get to

00:13:27.679 --> 00:13:31.959
something that looks like what

00:13:30.279 --> 00:13:34.439
these people run as a

00:13:31.960 --> 00:13:38.200
they run a regression essentially

00:13:34.440 --> 00:13:40.400
or they correlated and inflation

00:13:38.200 --> 00:13:42.720
inflation with unemployment

00:13:40.399 --> 00:13:44.240
and they found a downward sloping

00:13:42.720 --> 00:13:45.879
relationship

00:13:44.240 --> 00:13:47.839
well

00:13:45.879 --> 00:13:48.679
look what if that happens here suppose

00:13:47.839 --> 00:13:51.360
that

00:13:48.679 --> 00:13:54.599
that we assume that expected inflation

00:13:51.360 --> 00:13:56.480
is equal to some constant

00:13:54.600 --> 00:13:58.000
in in economics we say when that's the

00:13:56.480 --> 00:14:00.480
case

00:13:58.000 --> 00:14:03.679
and especially if pi is a low number

00:14:00.480 --> 00:14:06.159
inflation expectations are well anchored

00:14:03.679 --> 00:14:07.838
meaning you know any single year they

00:14:06.159 --> 00:14:09.159
can be a price of oil is high or

00:14:07.839 --> 00:14:10.720
something happens and inflation will

00:14:09.159 --> 00:14:12.600
deviate from that

00:14:10.720 --> 00:14:15.200
but people are all the time expecting

00:14:12.600 --> 00:14:17.000
for inflation sort of to go back to the

00:14:15.200 --> 00:14:18.560
what is a normal level

00:14:17.000 --> 00:14:19.799
nowadays

00:14:18.559 --> 00:14:22.679
or at least

00:14:19.799 --> 00:14:25.519
a few years ago in the US the normal

00:14:22.679 --> 00:14:28.719
level was around 2% say okay so people

00:14:25.519 --> 00:14:31.519
say well this year inflation was 1.8 but

00:14:28.720 --> 00:14:33.879
we expect next year 2% the next year we

00:14:31.519 --> 00:14:35.879
got surprises on the upside price of

00:14:33.879 --> 00:14:38.838
food went up something like that we got

00:14:35.879 --> 00:14:40.200
inflation of 2.3% but you ask people how

00:14:38.839 --> 00:14:42.240
do you what much do you expect for next

00:14:40.200 --> 00:14:45.400
year they say well 2%

00:14:42.240 --> 00:14:47.320
so so that's what a model of expectation

00:14:45.399 --> 00:14:49.679
like this means is that you know you're

00:14:47.320 --> 00:14:51.800
always expecting something which is

00:14:49.679 --> 00:14:54.000
some historical value on that we have

00:14:51.799 --> 00:14:56.199
agreed is a reasonable level for our

00:14:54.000 --> 00:14:58.759
economy or something like that okay

00:14:56.200 --> 00:15:01.879
so you see that if I replace expected

00:14:58.759 --> 00:15:04.000
inflation for a constant here pi bar

00:15:01.879 --> 00:15:06.720
then I have then my Phillips curve is

00:15:04.000 --> 00:15:09.120
really this is inflation then I have a

00:15:06.720 --> 00:15:11.120
constant minus alpha u

00:15:09.120 --> 00:15:12.879
that's the simplest of the

00:15:11.120 --> 00:15:14.919
downward sloping relationship I can have

00:15:12.879 --> 00:15:17.399
that case is a line downward sloping

00:15:14.919 --> 00:15:19.439
line no that's it

00:15:17.399 --> 00:15:20.799
of course you know it could be

00:15:19.440 --> 00:15:22.280
non-linear and so on but but this

00:15:20.799 --> 00:15:24.479
captures the essence so that's the

00:15:22.279 --> 00:15:27.919
theory for why

00:15:24.480 --> 00:15:29.600
Phillips was finding what he was finding

00:15:27.919 --> 00:15:31.719
our theory of the

00:15:29.600 --> 00:15:33.240
wage of the labor market if you will and

00:15:31.720 --> 00:15:35.959
the price setting

00:15:33.240 --> 00:15:39.159
behavior of firms gives us a Phillips

00:15:35.958 --> 00:15:43.319
curve of the kind that he had in mind

00:15:39.159 --> 00:15:45.360
and if you look in the 60s in the US

00:15:43.320 --> 00:15:47.079
then you see this negative relationship

00:15:45.360 --> 00:15:49.399
that eventually become sort of became a

00:15:47.078 --> 00:15:51.599
steeper it wasn't linear like this it

00:15:49.399 --> 00:15:53.480
was a little convex but but it's

00:15:51.600 --> 00:15:55.920
downward sloping

00:15:53.480 --> 00:15:58.200
and in fact

00:15:55.919 --> 00:16:02.120
to some extent

00:15:58.200 --> 00:16:04.000
our own very our very own Bob Solo and

00:16:02.120 --> 00:16:06.360
Paul Samuelson

00:16:04.000 --> 00:16:08.440
were advising the US government at the

00:16:06.360 --> 00:16:10.839
time and they said well you know let's

00:16:08.440 --> 00:16:13.280
exploit this stuff a little

00:16:10.839 --> 00:16:14.640
we like to have lower unemployment we

00:16:13.279 --> 00:16:16.639
can live with a little less more

00:16:14.639 --> 00:16:18.360
inflation but we know there's a negative

00:16:16.639 --> 00:16:20.799
trade-off there's a negative trade-off

00:16:18.360 --> 00:16:22.440
between these two things okay so

00:16:20.799 --> 00:16:24.359
if we like to lower unemployment it's

00:16:22.440 --> 00:16:26.280
fine we get a little more of inflation

00:16:24.360 --> 00:16:29.399
and initially the deal was very good

00:16:26.279 --> 00:16:31.360
because this curve was very flat

00:16:29.399 --> 00:16:33.000
you see you could cut him unemployment a

00:16:31.360 --> 00:16:35.000
lot you can see the dates here cutting

00:16:33.000 --> 00:16:37.000
unemployment a lot and you're not

00:16:35.000 --> 00:16:39.519
getting a lot of inflation

00:16:37.000 --> 00:16:41.519
eventually the deal the deal turned into

00:16:39.519 --> 00:16:42.720
a much rather deal much more rather deal

00:16:41.519 --> 00:16:43.958
because

00:16:42.720 --> 00:16:45.399
then to lower a little bit more

00:16:43.958 --> 00:16:48.958
unemployment we start getting a lot more

00:16:45.399 --> 00:16:50.838
of inflation okay so people for a while

00:16:48.958 --> 00:16:53.159
you know were okay with this model

00:16:50.839 --> 00:16:54.959
assuming that inflation was low but when

00:16:53.159 --> 00:16:57.159
they realized that this thing was being

00:16:54.958 --> 00:16:59.399
exploited then they began to sort of

00:16:57.159 --> 00:17:01.199
change the expectations they made I

00:16:59.399 --> 00:17:03.679
think that's what we had here but but

00:17:01.200 --> 00:17:05.120
the the reason held pretty well

00:17:03.679 --> 00:17:07.319
during this time and again it became

00:17:05.119 --> 00:17:09.359
steeper and steeper as we

00:17:07.319 --> 00:17:12.799
pushed it more and more towards sort of

00:17:09.359 --> 00:17:14.198
very low levels of unemployment

00:17:12.799 --> 00:17:15.399
so that's the story but again there is

00:17:14.199 --> 00:17:17.000
your model of the Phillips curve and

00:17:15.400 --> 00:17:18.720
that

00:17:17.000 --> 00:17:21.439
it's a very very good model for the

00:17:18.720 --> 00:17:23.838
times where Phillips also estimated his

00:17:21.439 --> 00:17:25.800
his Phillips curve

00:17:23.838 --> 00:17:27.919
now

00:17:25.799 --> 00:17:30.799
if you sort of turn the page and look at

00:17:27.920 --> 00:17:32.920
the same data in the 70s

00:17:30.799 --> 00:17:33.799
look how it looks

00:17:32.920 --> 00:17:36.279
okay

00:17:33.799 --> 00:17:37.720
so from 1970 to 1995 that's the data you

00:17:36.279 --> 00:17:40.440
have there there's no negative

00:17:37.720 --> 00:17:41.400
relationship thing is all over the place

00:17:40.440 --> 00:17:43.600
okay

00:17:41.400 --> 00:17:44.840
so had Mr. Phillips been born a few

00:17:43.599 --> 00:17:47.839
years

00:17:44.839 --> 00:17:49.240
few decades later and had he estimated

00:17:47.839 --> 00:17:50.480
his regression he would have found

00:17:49.240 --> 00:17:52.759
nothing

00:17:50.480 --> 00:17:54.279
there would be no curve in his honor at

00:17:52.759 --> 00:17:55.359
least if he had run that same regression

00:17:54.279 --> 00:17:56.759
maybe he would have run a different

00:17:55.359 --> 00:17:57.959
regression but

00:17:56.759 --> 00:18:00.440
nothing

00:17:57.960 --> 00:18:03.039
okay so what happened

00:18:00.440 --> 00:18:04.880
well our theory can explain as well what

00:18:03.039 --> 00:18:07.519
happened there

00:18:04.880 --> 00:18:09.600
remember the theory is not that

00:18:07.519 --> 00:18:11.319
inflation is equal to a constant minus

00:18:09.599 --> 00:18:13.399
the

00:18:11.319 --> 00:18:14.879
minus

00:18:13.400 --> 00:18:16.360
alpha u

00:18:14.880 --> 00:18:18.760
the theory says

00:18:16.359 --> 00:18:20.519
this is a constant only if

00:18:18.759 --> 00:18:21.759
the model of expectation is this

00:18:20.519 --> 00:18:24.400
constant

00:18:21.759 --> 00:18:26.920
but if expectation is moving around

00:18:24.400 --> 00:18:29.519
or if anything in this constant is

00:18:26.920 --> 00:18:31.200
moving around then then there's another

00:18:29.519 --> 00:18:34.359
source of variation

00:18:31.200 --> 00:18:36.400
okay for example what happens suppose

00:18:34.359 --> 00:18:39.279
that you're here in 1965 and all of the

00:18:36.400 --> 00:18:41.320
sudden you get a the price of oil goes

00:18:39.279 --> 00:18:43.319
up a lot and I'm telling you capture the

00:18:41.319 --> 00:18:44.918
price of oil with an increase in m firms

00:18:43.319 --> 00:18:46.559
need to sort of mark up things more in

00:18:44.919 --> 00:18:48.000
order to cover higher energy energy

00:18:46.559 --> 00:18:50.839
costs

00:18:48.000 --> 00:18:52.640
well look at what m does m says that for

00:18:50.839 --> 00:18:54.759
any given level of unemployment now I

00:18:52.640 --> 00:18:57.240
get higher inflation that's what an oil

00:18:54.759 --> 00:18:58.480
shock does no you get an oil shock then

00:18:57.240 --> 00:18:59.559
for any given level of unemployment now

00:18:58.480 --> 00:19:00.640
you get find yourself with more

00:18:59.559 --> 00:19:02.359
inflation

00:19:00.640 --> 00:19:04.759
so that moves you in the opposite

00:19:02.359 --> 00:19:06.839
direction moves you up there and that's

00:19:04.759 --> 00:19:08.759
one of the reasons

00:19:06.839 --> 00:19:10.480
for these points around here

00:19:08.759 --> 00:19:12.158
we got lots of inflation because we got

00:19:10.480 --> 00:19:17.360
massive oil shocks

00:19:12.159 --> 00:19:17.360
during the 70s and early 80s okay

00:19:18.720 --> 00:19:23.559
we had wars in the Middle East and so on

00:19:20.599 --> 00:19:25.719
that that led to to those shocks so that

00:19:23.559 --> 00:19:27.960
so that was one of the reasons

00:19:25.720 --> 00:19:29.600
we got shocks here

00:19:27.960 --> 00:19:31.480
to this term here

00:19:29.599 --> 00:19:33.240
and that sort of

00:19:31.480 --> 00:19:35.159
muddied the relationship

00:19:33.240 --> 00:19:37.200
but the other reason which is more

00:19:35.159 --> 00:19:38.679
interesting I think and that you already

00:19:37.200 --> 00:19:40.600
began to see that something was

00:19:38.679 --> 00:19:41.640
happening here

00:19:40.599 --> 00:19:44.119
is that

00:19:41.640 --> 00:19:45.720
as inflation went up

00:19:44.119 --> 00:19:48.399
people sort of stopped believing in this

00:19:45.720 --> 00:19:52.200
model so the expectation formation

00:19:48.400 --> 00:19:52.200
mechanism changed

00:19:52.559 --> 00:19:56.039
okay so this guy

00:19:54.200 --> 00:19:57.120
began to react

00:19:56.039 --> 00:19:58.839
to

00:19:57.119 --> 00:20:01.519
endogenous variables. And I'm going to

00:19:58.839 --> 00:20:03.959
explain more precisely what So, that's

00:20:01.519 --> 00:20:06.680
what we mean by expected inflation

00:20:03.960 --> 00:20:09.480
became the anchor. It was no longer

00:20:06.680 --> 00:20:11.560
anchored around this constant of 2%

00:20:09.480 --> 00:20:13.720
but but but it became the anchor. It

00:20:11.559 --> 00:20:15.559
began to follow the data. So, if if the

00:20:13.720 --> 00:20:17.319
data came with more inflation, then

00:20:15.559 --> 00:20:19.559
people believed that next year we would

00:20:17.319 --> 00:20:22.799
have more inflation as well. Okay, not

00:20:19.559 --> 00:20:24.279
back to 2% but if we got 5% inflation

00:20:22.799 --> 00:20:26.359
today, people began to say, "Well, okay,

00:20:24.279 --> 00:20:28.359
I don't think that next year my best

00:20:26.359 --> 00:20:30.799
estimate is 2% is probably closer to

00:20:28.359 --> 00:20:33.079
5%." Okay, that's what it means

00:20:30.799 --> 00:20:34.919
de-anchoring. That's what has the Fed

00:20:33.079 --> 00:20:36.519
and most central banks around the world

00:20:34.920 --> 00:20:39.680
terrified today.

00:20:36.519 --> 00:20:42.319
Inflation is very is much higher than 2%

00:20:39.680 --> 00:20:43.440
and they're very worried about this guy

00:20:42.319 --> 00:20:46.319
becoming

00:20:43.440 --> 00:20:48.440
the anchor or an anchor. Okay.

00:20:46.319 --> 00:20:50.279
I'll get back to that in a second.

00:20:48.440 --> 00:20:51.200
Anyway, so but let me let me explain

00:20:50.279 --> 00:20:53.000
this

00:20:51.200 --> 00:20:54.279
how this expected inflation term work

00:20:53.000 --> 00:20:56.880
here.

00:20:54.279 --> 00:20:59.279
So, let me replace the model of

00:20:56.880 --> 00:21:00.560
expected inflation

00:20:59.279 --> 00:21:02.639
for something which is some weighted

00:21:00.559 --> 00:21:03.720
average of a constant.

00:21:02.640 --> 00:21:05.120
That's

00:21:03.720 --> 00:21:07.519
and

00:21:05.119 --> 00:21:09.879
the most recent inflation.

00:21:07.519 --> 00:21:12.079
Okay. So, this model says, "What is my

00:21:09.880 --> 00:21:13.800
expected inflation for next year? Well,

00:21:12.079 --> 00:21:15.279
it's an average of this long-run target

00:21:13.799 --> 00:21:17.000
that we have,

00:21:15.279 --> 00:21:18.279
say 2%,

00:21:17.000 --> 00:21:20.799
and whatever was the most recent

00:21:18.279 --> 00:21:22.559
inflation."

00:21:20.799 --> 00:21:24.839
If theta in the model I showed you

00:21:22.559 --> 00:21:25.960
before, the one that applied to the '60s

00:21:24.839 --> 00:21:28.399
and so on,

00:21:25.960 --> 00:21:30.000
up to the '60s, had essentially theta

00:21:28.400 --> 00:21:32.920
equal to zero.

00:21:30.000 --> 00:21:35.400
So, this guy didn't show up there

00:21:32.920 --> 00:21:36.880
and and expected inflation was very well

00:21:35.400 --> 00:21:39.440
anchored.

00:21:36.880 --> 00:21:41.880
What we began to happen

00:21:39.440 --> 00:21:43.720
as we began to move that way

00:21:41.880 --> 00:21:45.240
and then we got hit by oil shocks so we

00:21:43.720 --> 00:21:46.920
So, people began to see much higher

00:21:45.240 --> 00:21:48.039
inflation numbers than they were used

00:21:46.920 --> 00:21:50.120
to,

00:21:48.039 --> 00:21:51.240
then this theta

00:21:50.119 --> 00:21:54.039
began to

00:21:51.240 --> 00:21:55.880
increase. Okay. So, people began to sort

00:21:54.039 --> 00:21:57.279
of change the model of expectation and

00:21:55.880 --> 00:21:58.560
began to think that

00:21:57.279 --> 00:22:00.519
inflation was going to be more

00:21:58.559 --> 00:22:02.399
persistent than they used to think in

00:22:00.519 --> 00:22:03.920
the past. So, in high inflation today

00:22:02.400 --> 00:22:05.800
means high inflation tomorrow. That's

00:22:03.920 --> 00:22:07.920
what it means more persistent. In the

00:22:05.799 --> 00:22:10.240
past was high inflation today was was a

00:22:07.920 --> 00:22:12.920
back draw, we'll go back to sort of the

00:22:10.240 --> 00:22:15.039
normal long-run average. Now, that's no

00:22:12.920 --> 00:22:16.640
longer the case. And so, if I replace

00:22:15.039 --> 00:22:18.119
this more general model of expected

00:22:16.640 --> 00:22:20.280
inflation

00:22:18.119 --> 00:22:22.359
here in the Phillips curve, I get this

00:22:20.279 --> 00:22:23.440
expression which now has this extra

00:22:22.359 --> 00:22:24.678
term.

00:22:23.440 --> 00:22:26.640
So, the

00:22:24.679 --> 00:22:29.080
we used to have theta equal to zero but

00:22:26.640 --> 00:22:31.480
during the '70s and '80s and even early

00:22:29.079 --> 00:22:33.879
'90s, actually that theta got to be very

00:22:31.480 --> 00:22:35.799
close to one.

00:22:33.880 --> 00:22:37.920
Okay, if you estimate these models, you

00:22:35.799 --> 00:22:39.678
get that theta was very close to one.

00:22:37.920 --> 00:22:41.200
And look at what happens when theta gets

00:22:39.679 --> 00:22:43.759
very close to one.

00:22:41.200 --> 00:22:45.519
So, when theta is is one literally, then

00:22:43.759 --> 00:22:47.400
the best forecast for inflation is the

00:22:45.519 --> 00:22:49.240
previous inflation.

00:22:47.400 --> 00:22:53.400
Okay, so this year is 5% and I think

00:22:49.240 --> 00:22:55.240
next year is 5%, not 2%, 5%.

00:22:53.400 --> 00:22:57.759
If this year is 7%, I think next year is

00:22:55.240 --> 00:23:00.319
7% again.

00:22:57.759 --> 00:23:02.440
And so, if you do that, then my expected

00:23:00.319 --> 00:23:04.879
inflation becomes lag inflation pi t

00:23:02.440 --> 00:23:07.679
minus one. So, if I stick in replace the

00:23:04.880 --> 00:23:09.520
expected inflation for pi t minus one, I

00:23:07.679 --> 00:23:11.320
get to this Phillips curve

00:23:09.519 --> 00:23:14.920
which I can rewrite

00:23:11.319 --> 00:23:16.960
as the change in inflation in relation

00:23:14.920 --> 00:23:18.840
as as a relationship between the change

00:23:16.960 --> 00:23:20.319
in inflation and the level of

00:23:18.839 --> 00:23:22.240
unemployment.

00:23:20.319 --> 00:23:25.639
So, now what you have is that if

00:23:22.240 --> 00:23:28.319
unemployment is very low, then inflation

00:23:25.640 --> 00:23:30.600
is is picking up, you know, it's going

00:23:28.319 --> 00:23:33.279
there. So, if inflation if unemployment

00:23:30.599 --> 00:23:34.719
is very low, not only inflation is high,

00:23:33.279 --> 00:23:36.200
but it's also growing

00:23:34.720 --> 00:23:37.319
over time.

00:23:36.200 --> 00:23:39.440
Okay.

00:23:37.319 --> 00:23:41.279
That's the reason sometimes people refer

00:23:39.440 --> 00:23:44.200
to this formulation of the Phillips

00:23:41.279 --> 00:23:45.720
curve as the accelerationist Phillips

00:23:44.200 --> 00:23:47.759
curve because now there's a relation

00:23:45.720 --> 00:23:50.559
between unemployment and the change in

00:23:47.759 --> 00:23:51.879
inflation. And if you estimate

00:23:50.559 --> 00:23:54.200
this Phillips curve, this

00:23:51.880 --> 00:23:57.160
accelerationist Phillips curve on the

00:23:54.200 --> 00:23:59.000
data I just showed you of the '70s and

00:23:57.160 --> 00:24:00.880
'80s, you get a much better

00:23:59.000 --> 00:24:03.119
relationship. Okay, you still have the

00:24:00.880 --> 00:24:05.120
oil shocks that messed things up

00:24:03.119 --> 00:24:07.319
but but you can start seeing recovering

00:24:05.119 --> 00:24:09.079
this negative relationship. But again,

00:24:07.319 --> 00:24:10.559
it's between the change in inflation and

00:24:09.079 --> 00:24:12.480
the level of unemployment. And that's a

00:24:10.559 --> 00:24:14.599
very scary situation for a central bank

00:24:12.480 --> 00:24:17.839
to find itself in because it's very easy

00:24:14.599 --> 00:24:17.839
to for things to escalate.

00:24:19.160 --> 00:24:21.320
Okay.

00:24:22.359 --> 00:24:27.559
So, by the mid-'90s,

00:24:25.160 --> 00:24:30.200
we had re-anchored expectations. There

00:24:27.559 --> 00:24:32.119
was a sort of very aggressive

00:24:30.200 --> 00:24:33.920
policy to control inflation by Paul

00:24:32.119 --> 00:24:37.359
Volcker

00:24:33.920 --> 00:24:39.200
in the US and it was imitated around the

00:24:37.359 --> 00:24:41.719
world with some lag,

00:24:39.200 --> 00:24:43.880
but but inflation became re-anchored.

00:24:41.720 --> 00:24:46.440
So, we went back to this theta equal to

00:24:43.880 --> 00:24:48.000
zero type model. The expected inflation

00:24:46.440 --> 00:24:51.200
in the US, the target inflation of the

00:24:48.000 --> 00:24:53.759
central bank was around 2% that became

00:24:51.200 --> 00:24:56.200
what people expected for the next year

00:24:53.759 --> 00:24:59.400
and and that re-anchored. So, we went

00:24:56.200 --> 00:24:59.400
back in other words

00:25:00.279 --> 00:25:04.920
to that sort of Phillips curve.

00:25:03.160 --> 00:25:06.960
Okay, and that's what central banks want

00:25:04.920 --> 00:25:09.039
to be at. They want to have inflation

00:25:06.960 --> 00:25:10.519
expectation very well anchored.

00:25:09.039 --> 00:25:12.960
And they were very successful after the

00:25:10.519 --> 00:25:15.639
'90s. And so, we got

00:25:12.960 --> 00:25:17.360
into again now Look, I'm I'm now I'm not

00:25:15.640 --> 00:25:19.440
running the accelerations. I'm again

00:25:17.359 --> 00:25:21.359
running inflation against unemployment

00:25:19.440 --> 00:25:24.320
and you again can see this downward

00:25:21.359 --> 00:25:26.000
sloping relationship. Okay.

00:25:24.319 --> 00:25:28.159
So, that was very good news. Was great

00:25:26.000 --> 00:25:31.319
success of monetary policy

00:25:28.160 --> 00:25:34.080
during the '90s and later was the

00:25:31.319 --> 00:25:36.159
re-anchoring of expected inflation again

00:25:34.079 --> 00:25:37.960
all around the developed world and many

00:25:36.160 --> 00:25:39.720
of the include even

00:25:37.960 --> 00:25:41.559
Latin America, many economies in Latin

00:25:39.720 --> 00:25:43.400
America saw re-anchored expectation,

00:25:41.559 --> 00:25:47.159
Asia and so on. So,

00:25:43.400 --> 00:25:47.160
so it was a good time for central banks.

00:25:51.759 --> 00:25:54.279
Okay.

00:25:59.039 --> 00:26:01.799
So, the next

00:26:00.119 --> 00:26:03.639
thing I want to do, this will connect

00:26:01.799 --> 00:26:05.200
more with the with the the previous

00:26:03.640 --> 00:26:07.480
lecture. This is the last thing I want

00:26:05.200 --> 00:26:10.240
to say

00:26:07.480 --> 00:26:12.160
for this lecture then I'm I may start

00:26:10.240 --> 00:26:13.920
the review afterwards.

00:26:12.160 --> 00:26:16.160
Um

00:26:13.920 --> 00:26:17.759
is that I want to connect now this

00:26:16.160 --> 00:26:19.160
Phillips curve with something we

00:26:17.759 --> 00:26:20.599
discussed in the previous lecture which

00:26:19.160 --> 00:26:22.120
is the natural rate of unemployment

00:26:20.599 --> 00:26:23.839
because that's the way

00:26:22.119 --> 00:26:26.039
you'll typically see the Phillips curve

00:26:23.839 --> 00:26:27.599
written and and that's

00:26:26.039 --> 00:26:30.559
also the way

00:26:27.599 --> 00:26:32.480
that sort of uh you know, when Chairman

00:26:30.559 --> 00:26:34.839
Powell is talking about the labor market

00:26:32.480 --> 00:26:37.039
tightness and so on, he's not talking

00:26:34.839 --> 00:26:38.319
relative to M and Z and things like

00:26:37.039 --> 00:26:39.960
that, he's talking relative to what is

00:26:38.319 --> 00:26:42.119
called the natural rate of unemployment.

00:26:39.960 --> 00:26:45.360
So, I want to go from a Phillips curve

00:26:42.119 --> 00:26:45.359
that looks like that,

00:26:46.119 --> 00:26:50.799
you know, like that, to one that has the

00:26:48.400 --> 00:26:53.080
natural rate of unemployment in there.

00:26:50.799 --> 00:26:54.799
And so, that's the last step

00:26:53.079 --> 00:26:55.759
in this lecture.

00:26:54.799 --> 00:26:57.240
So,

00:26:55.759 --> 00:26:59.039
remember the definition of the natural

00:26:57.240 --> 00:27:00.160
rate of unemployment. What was the

00:26:59.039 --> 00:27:02.839
definition of the natural rate of

00:27:00.160 --> 00:27:02.840
unemployment?

00:27:04.400 --> 00:27:09.960
Was it the unemployment rate that God

00:27:07.400 --> 00:27:09.960
gave us?

00:27:10.679 --> 00:27:13.240
Any God?

00:27:15.480 --> 00:27:21.079
No.

00:27:17.480 --> 00:27:21.079
It had a very precise meaning for us.

00:27:24.000 --> 00:27:29.920
And remember, we used exactly that model

00:27:27.119 --> 00:27:29.919
to figure it out.

00:27:30.319 --> 00:27:32.559
Remember?

00:27:34.240 --> 00:27:38.120
We we solved the Actually, we solved the

00:27:36.679 --> 00:27:39.800
natural rate of unemployment from

00:27:38.119 --> 00:27:41.959
something like this. I think we had the

00:27:39.799 --> 00:27:44.119
function still generic function f of U

00:27:41.960 --> 00:27:46.120
Z. But we solved from an expression like

00:27:44.119 --> 00:27:47.919
this.

00:27:46.119 --> 00:27:50.119
We said,

00:27:47.920 --> 00:27:51.679
"Under one assumption,

00:27:50.119 --> 00:27:52.839
we can call this

00:27:51.679 --> 00:27:55.200
U

00:27:52.839 --> 00:27:57.119
U N, the natural rate of unemployment.

00:27:55.200 --> 00:27:59.840
What was that assumption?"

00:27:57.119 --> 00:28:01.839
And that's the only thing

00:27:59.839 --> 00:28:03.639
Expected price Okay, expected price is

00:28:01.839 --> 00:28:05.678
equal to the actual price. Okay, so we

00:28:03.640 --> 00:28:07.440
said if this is equal to that, then you

00:28:05.679 --> 00:28:08.800
solve out that's the natural rate of

00:28:07.440 --> 00:28:12.360
unemployment. And that's the only thing

00:28:08.799 --> 00:28:14.000
that that it that means that that that

00:28:12.359 --> 00:28:15.919
natural rate of unemployment means

00:28:14.000 --> 00:28:17.400
simply that when when the

00:28:15.920 --> 00:28:19.840
when the price is equal to the expected

00:28:17.400 --> 00:28:19.840
price.

00:28:20.319 --> 00:28:23.279
But if the price

00:28:23.839 --> 00:28:29.799
is equal to the expected price,

00:28:26.960 --> 00:28:29.799
what else is equal?

00:28:33.079 --> 00:28:37.119
I pointed at the right expressions

00:28:35.000 --> 00:28:37.119
there.

00:28:37.559 --> 00:28:41.000
Inflation is equal to expected

00:28:39.279 --> 00:28:44.599
inflation.

00:28:41.000 --> 00:28:45.799
So, I can use the same logic I used here

00:28:44.599 --> 00:28:47.799
for the natural rate of unemployment

00:28:45.799 --> 00:28:49.519
using the Phillips curve.

00:28:47.799 --> 00:28:51.440
I can say, "Okay,

00:28:49.519 --> 00:28:53.799
my I can solve out for the natural rate

00:28:51.440 --> 00:28:55.679
of unemployment here simply by setting

00:28:53.799 --> 00:28:58.279
the expected inflation equal to actual

00:28:55.679 --> 00:28:58.280
inflation."

00:28:58.640 --> 00:29:01.679
Okay.

00:28:59.720 --> 00:29:04.039
And if I do this, I can solve for the

00:29:01.679 --> 00:29:05.159
natural rate of unemployment from here.

00:29:04.039 --> 00:29:06.599
U N.

00:29:05.159 --> 00:29:08.480
I mean, I'm going to give I'm going to

00:29:06.599 --> 00:29:11.359
put the superscript N here when I when

00:29:08.480 --> 00:29:12.319
you let me replace pi e for pi. That's a

00:29:11.359 --> 00:29:13.918
That's what I

00:29:12.319 --> 00:29:16.119
That's what I

00:29:13.919 --> 00:29:17.840
The fact that I replace this pi e for pi

00:29:16.119 --> 00:29:19.519
is what allows me to put the superscript

00:29:17.839 --> 00:29:21.279
N there. Call it the natural rate of

00:29:19.519 --> 00:29:22.720
unemployment. And now I can solve it.

00:29:21.279 --> 00:29:23.960
Well, obviously that cancels with that

00:29:22.720 --> 00:29:25.960
and I can solve the natural rate of

00:29:23.960 --> 00:29:28.720
unemployment and it's equal to this

00:29:25.960 --> 00:29:30.279
function here.

00:29:28.720 --> 00:29:33.440
So, why is the natural rate of

00:29:30.279 --> 00:29:35.759
unemployment increasing in M?

00:29:33.440 --> 00:29:38.080
A question like that can come up in the

00:29:35.759 --> 00:29:39.480
quiz.

00:29:38.079 --> 00:29:41.039
I'm not going to use the Phillips curve

00:29:39.480 --> 00:29:42.640
to ask you if I ask you about that, but

00:29:41.039 --> 00:29:43.759
I can ask you that. What

00:29:42.640 --> 00:29:47.440
What happens to the natural rate of

00:29:43.759 --> 00:29:47.440
unemployment if M goes up?

00:29:47.519 --> 00:29:50.680
You know that.

00:29:49.119 --> 00:29:53.319
UN will go up, but what is the

00:29:50.680 --> 00:29:53.320
mechanism?

00:29:59.200 --> 00:30:02.880
So, why does the natural rate of

00:30:00.440 --> 00:30:06.039
unemployment go up when

00:30:02.880 --> 00:30:08.000
the markup goes up?

00:30:06.039 --> 00:30:10.039
Yep. If the real cost is constant, wages

00:30:08.000 --> 00:30:12.039
have to go down, right?

00:30:10.039 --> 00:30:14.000
I mean, another way of saying it is that

00:30:12.039 --> 00:30:16.799
the firms are not willing to pay they

00:30:14.000 --> 00:30:19.759
want to pay a lower real wage.

00:30:16.799 --> 00:30:22.159
At the original level of unemployment

00:30:19.759 --> 00:30:24.319
before the change in M,

00:30:22.160 --> 00:30:26.279
workers would not take that lower real

00:30:24.319 --> 00:30:27.879
wage.

00:30:26.279 --> 00:30:29.319
No, it's not an equilibrium real wage

00:30:27.880 --> 00:30:31.000
because workers say, "No, no, at this

00:30:29.319 --> 00:30:32.879
level of unemployment we need a higher

00:30:31.000 --> 00:30:34.559
real wage."

00:30:32.880 --> 00:30:37.120
So, the only way to restore equilibrium

00:30:34.559 --> 00:30:39.039
in that model we had was to increase

00:30:37.119 --> 00:30:40.959
unemployment because that will lower the

00:30:39.039 --> 00:30:42.920
bargaining power of workers and they

00:30:40.960 --> 00:30:44.720
will end up accepting the lower real

00:30:42.920 --> 00:30:46.320
wage that firms are willing to offer

00:30:44.720 --> 00:30:48.039
now. Okay.

00:30:46.319 --> 00:30:52.319
So, that's the reason

00:30:48.039 --> 00:30:52.319
uh we get this this markup effect.

00:30:53.359 --> 00:30:57.479
Z is

00:30:55.279 --> 00:31:00.079
same logic. It's a little easier to see

00:30:57.480 --> 00:31:02.400
it there, but Z means, well, at any

00:31:00.079 --> 00:31:04.639
given level of unemployment

00:31:02.400 --> 00:31:06.320
an increase in Z means workers want a

00:31:04.640 --> 00:31:07.840
higher real wage.

00:31:06.319 --> 00:31:09.159
Firms are not willing to pay a higher

00:31:07.839 --> 00:31:11.119
real wage,

00:31:09.160 --> 00:31:13.200
so you have to bring down the real wage

00:31:11.119 --> 00:31:14.399
that workers demand and the only way

00:31:13.200 --> 00:31:15.680
that can happen is with a higher

00:31:14.400 --> 00:31:17.000
unemployment.

00:31:15.680 --> 00:31:22.120
Okay. That's the reason the natural rate

00:31:17.000 --> 00:31:22.119
of unemployment is also increasing in Z.

00:31:23.119 --> 00:31:26.799
Okay.

00:31:24.440 --> 00:31:29.559
And now the last step.

00:31:26.799 --> 00:31:31.319
The last step is to

00:31:29.559 --> 00:31:33.799
You see, I can go back to my Phillips

00:31:31.319 --> 00:31:33.799
curve.

00:31:34.400 --> 00:31:40.480
Say that.

00:31:36.920 --> 00:31:45.080
And I'm going to replace M plus Z

00:31:40.480 --> 00:31:48.839
for alpha UN. I can do that, you see?

00:31:45.079 --> 00:31:51.919
I can replace this M plus C Z for alpha

00:31:48.839 --> 00:31:51.919
times UN.

00:31:52.680 --> 00:31:58.679
How do I know that? Well, M plus C Z is

00:31:55.559 --> 00:32:01.039
equal to UN times alpha.

00:31:58.679 --> 00:32:04.800
I can replace in the Phillips curve

00:32:01.039 --> 00:32:06.319
M plus C by alpha UN and I can re

00:32:04.799 --> 00:32:08.000
I can therefore

00:32:06.319 --> 00:32:10.159
rewrite the Phillips curve in the

00:32:08.000 --> 00:32:12.319
following form.

00:32:10.160 --> 00:32:16.200
Inflation is equal to expected inflation

00:32:12.319 --> 00:32:18.359
minus alpha times the gap between the

00:32:16.200 --> 00:32:20.519
unemployment rate and the natural rate

00:32:18.359 --> 00:32:22.359
of unemployment.

00:32:20.519 --> 00:32:23.599
Okay. So,

00:32:22.359 --> 00:32:25.279
so

00:32:23.599 --> 00:32:27.759
when

00:32:25.279 --> 00:32:29.519
Chairman Powell is worried about labor

00:32:27.759 --> 00:32:31.879
market being very tight, what he's

00:32:29.519 --> 00:32:33.240
saying is, well, unemployment is likely

00:32:31.880 --> 00:32:34.920
to be below the natural rate of

00:32:33.240 --> 00:32:36.559
unemployment.

00:32:34.920 --> 00:32:37.840
Because if unemployment is below the

00:32:36.559 --> 00:32:41.359
natural rate of unemployment, that's

00:32:37.839 --> 00:32:43.720
putting upward pressure on inflation.

00:32:41.359 --> 00:32:43.719
Okay.

00:32:44.160 --> 00:32:47.160
So, that's a

00:32:45.359 --> 00:32:50.559
So, that's what it means. This gap is

00:32:47.160 --> 00:32:52.040
very important uh for macroeconomists

00:32:50.559 --> 00:32:54.480
and certainly for central bankers that

00:32:52.039 --> 00:32:55.720
are very worried about inflation. Okay?

00:32:54.480 --> 00:32:58.120
That gap here.

00:32:55.720 --> 00:32:59.880
Problem is is this this is a difficult

00:32:58.119 --> 00:33:02.119
object to estimate, so you have to have

00:32:59.880 --> 00:33:04.360
estimates as

00:33:02.119 --> 00:33:05.719
The truth is that it's very difficult to

00:33:04.359 --> 00:33:06.959
know what it is, although there are

00:33:05.720 --> 00:33:09.039
estimates out there and I'm going to

00:33:06.960 --> 00:33:10.559
show you one.

00:33:09.039 --> 00:33:12.759
You notice that something is wrong when

00:33:10.559 --> 00:33:14.159
this guy starts picking up. It's a It's

00:33:12.759 --> 00:33:15.160
a little bit the other way around, you

00:33:14.160 --> 00:33:16.679
know?

00:33:15.160 --> 00:33:18.880
Uh uh

00:33:16.679 --> 00:33:20.759
the US in fact had a the opposite

00:33:18.880 --> 00:33:21.840
problem

00:33:20.759 --> 00:33:23.720
um

00:33:21.839 --> 00:33:25.399
before COVID. It's a somehow

00:33:23.720 --> 00:33:27.279
unemployment was very low relative to

00:33:25.400 --> 00:33:28.640
historical levels, but inflation was not

00:33:27.279 --> 00:33:30.759
picking up.

00:33:28.640 --> 00:33:33.360
So, that was implicitly telling us that

00:33:30.759 --> 00:33:34.839
for some reason, not fully understood,

00:33:33.359 --> 00:33:36.479
the natural rate of unemployment was

00:33:34.839 --> 00:33:38.199
declining.

00:33:36.480 --> 00:33:40.880
Okay.

00:33:38.200 --> 00:33:43.519
So, here is one picture that looks

00:33:40.880 --> 00:33:45.840
is one estimate uh again, I I don't

00:33:43.519 --> 00:33:48.400
trust any particular estimate, but

00:33:45.839 --> 00:33:49.559
it tells a story. That's one particular

00:33:48.400 --> 00:33:52.360
estimate of the natural rate of

00:33:49.559 --> 00:33:54.480
unemployment in the US, that blue line.

00:33:52.359 --> 00:33:57.519
And what you see in red the red is the

00:33:54.480 --> 00:33:58.799
actual rate of unemployment in the US.

00:33:57.519 --> 00:34:00.559
So,

00:33:58.799 --> 00:34:03.159
what happens when when in situations

00:34:00.559 --> 00:34:03.159
like these?

00:34:04.640 --> 00:34:08.240
So, what do you think what's happening

00:34:05.960 --> 00:34:09.679
to inflation in in this episode, which

00:34:08.239 --> 00:34:12.959
is right after the global financial

00:34:09.679 --> 00:34:12.960
crisis or the great recession?

00:34:15.480 --> 00:34:18.398
So, what what what do you need to read

00:34:17.039 --> 00:34:19.960
here? Well,

00:34:18.398 --> 00:34:21.839
the unemployment rate was a lot higher

00:34:19.960 --> 00:34:24.159
than the

00:34:21.840 --> 00:34:25.760
natural rate of unemployment.

00:34:24.159 --> 00:34:28.159
Does that put upward or downward

00:34:25.760 --> 00:34:29.440
pressure on inflation?

00:34:28.159 --> 00:34:30.800
Downward pressure on inflation. No,

00:34:29.440 --> 00:34:32.840
unemployment is very high relative to

00:34:30.800 --> 00:34:34.960
natural rate of unemployment. It's minus

00:34:32.840 --> 00:34:36.440
alpha times U minus UN.

00:34:34.960 --> 00:34:37.878
So, and that's what happened. We had

00:34:36.440 --> 00:34:39.599
lots of problem with inflation.

00:34:37.878 --> 00:34:41.519
Inflation was going very low. We even

00:34:39.599 --> 00:34:43.079
had negative inflation there, a little

00:34:41.519 --> 00:34:46.800
deflation for a while.

00:34:43.079 --> 00:34:48.358
Okay. So, that was a problem.

00:34:46.800 --> 00:34:49.879
Here is the period that they described

00:34:48.358 --> 00:34:51.759
before is a little mysterious because we

00:34:49.878 --> 00:34:52.719
went unemployment went below what we

00:34:51.760 --> 00:34:54.120
thought it was a natural rate of

00:34:52.719 --> 00:34:55.878
unemployment and inflation wasn't really

00:34:54.119 --> 00:34:57.679
picking up a lot. At the end began to

00:34:55.878 --> 00:34:59.119
pick up a little, but it wasn't picking

00:34:57.679 --> 00:35:00.839
up a lot and that was a little bit of a

00:34:59.119 --> 00:35:03.319
mystery.

00:35:00.840 --> 00:35:04.519
Now, we're in this situation here,

00:35:03.320 --> 00:35:07.760
which

00:35:04.519 --> 00:35:10.519
we have extremely low unemployment

00:35:07.760 --> 00:35:12.160
and very high inflation. So, so this I

00:35:10.519 --> 00:35:13.639
think this captures well the situation

00:35:12.159 --> 00:35:15.599
right now. We have a

00:35:13.639 --> 00:35:17.639
negative gap between unemployment and

00:35:15.599 --> 00:35:18.759
the natural rate of unemployment and

00:35:17.639 --> 00:35:20.759
that's the reason that's putting a lot

00:35:18.760 --> 00:35:22.160
of pressure on inflation.

00:35:20.760 --> 00:35:23.520
We also have other things that are

00:35:22.159 --> 00:35:25.079
putting pressure on inflation that come

00:35:23.519 --> 00:35:26.000
from the supply side of the economy and

00:35:25.079 --> 00:35:29.440
so on.

00:35:26.000 --> 00:35:31.639
So, that combination is pretty bad for

00:35:29.440 --> 00:35:35.039
for the

00:35:31.639 --> 00:35:36.079
inflation outcomes and outlook

00:35:35.039 --> 00:35:38.199
as well.

00:35:36.079 --> 00:35:38.199
Okay.

00:35:38.639 --> 00:35:41.960
So, that's where we're at.

00:35:40.800 --> 00:35:43.680
We're going to talk a lot more about

00:35:41.960 --> 00:35:45.440
this because this is what is going on

00:35:43.679 --> 00:35:47.000
right now.

00:35:45.440 --> 00:35:48.559
Any questions about that? Otherwise, I

00:35:47.000 --> 00:35:51.358
want to start sort of reviewing things,

00:35:48.559 --> 00:35:51.358
although I don't know.

00:35:51.599 --> 00:35:57.319
Any question about this? Yep.

00:35:54.840 --> 00:35:59.600
Is correction to increase unemployment?

00:35:57.320 --> 00:36:01.320
Sorry? Is the only way to fix, I guess,

00:35:59.599 --> 00:36:03.880
the inflationary expectations? Well,

00:36:01.320 --> 00:36:07.280
that's a very good question.

00:36:03.880 --> 00:36:07.280
That's a very good question.

00:36:09.119 --> 00:36:13.000
I'm I'm trying to decide what to

00:36:14.039 --> 00:36:20.320
answer what with what do we have.

00:36:17.719 --> 00:36:20.319
Um

00:36:23.239 --> 00:36:27.039
There are two views

00:36:24.880 --> 00:36:29.800
at this moment.

00:36:27.039 --> 00:36:32.440
There's one view

00:36:29.800 --> 00:36:33.760
that says there's no way around that.

00:36:32.440 --> 00:36:34.679
They just look at these curves and say,

00:36:33.760 --> 00:36:35.800
"Look,

00:36:34.679 --> 00:36:38.039
there's no way around that. That's the

00:36:35.800 --> 00:36:38.920
reason we need a recession."

00:36:38.039 --> 00:36:40.440
Okay.

00:36:38.920 --> 00:36:43.280
Because otherwise we were not going to

00:36:40.440 --> 00:36:45.639
control inflation.

00:36:43.280 --> 00:36:47.760
And a recession means high unemployment.

00:36:45.639 --> 00:36:50.079
Okay, that's one view.

00:36:47.760 --> 00:36:51.400
At this moment, it's becoming the

00:36:50.079 --> 00:36:53.079
dominant view.

00:36:51.400 --> 00:36:56.320
It has gone in cycles, but at this

00:36:53.079 --> 00:36:56.319
moment it's the dominant view.

00:36:57.000 --> 00:37:01.840
There is a another view,

00:36:59.800 --> 00:37:04.320
which is the one that the central bank

00:37:01.840 --> 00:37:06.400
the Fed adopted for a while,

00:37:04.320 --> 00:37:08.160
that said, "Well, this is not the only

00:37:06.400 --> 00:37:11.200
indicator of tightness of the labor

00:37:08.159 --> 00:37:13.239
market. There is other things as well."

00:37:11.199 --> 00:37:14.879
And those indicators are moving in the

00:37:13.239 --> 00:37:16.799
right direction.

00:37:14.880 --> 00:37:18.920
And so, we may be able not to create a

00:37:16.800 --> 00:37:21.080
big mess here because these other

00:37:18.920 --> 00:37:22.680
factors are moving in the right right

00:37:21.079 --> 00:37:24.119
direction.

00:37:22.679 --> 00:37:26.079
Some of those factors are as I said,

00:37:24.119 --> 00:37:28.400
other measures of of labor market

00:37:26.079 --> 00:37:29.799
tightness and and hiring, the flows.

00:37:28.400 --> 00:37:31.160
Remember I showed you flows between

00:37:29.800 --> 00:37:33.240
employment and unemployment, out of

00:37:31.159 --> 00:37:35.000
employment and so on. Those flows look

00:37:33.239 --> 00:37:36.799
extremely tight and now they're

00:37:35.000 --> 00:37:38.920
improving. So, the gaps in those

00:37:36.800 --> 00:37:40.960
dimensions are better. And the other one

00:37:38.920 --> 00:37:42.559
is the what's a big cost push component,

00:37:40.960 --> 00:37:44.519
which is what I said before, the supply

00:37:42.559 --> 00:37:46.320
chains and so on created extra

00:37:44.519 --> 00:37:48.400
inflation, abnormal inflation like

00:37:46.320 --> 00:37:49.280
increasing markups, like M was very

00:37:48.400 --> 00:37:50.800
high.

00:37:49.280 --> 00:37:53.040
And some of that is subsiding as well.

00:37:50.800 --> 00:37:54.800
So, so there are dynamics that suggest

00:37:53.039 --> 00:37:56.119
that inflation is declining even without

00:37:54.800 --> 00:37:58.160
unemployment.

00:37:56.119 --> 00:38:00.039
But, I would say

00:37:58.159 --> 00:38:02.519
the medium voter

00:38:00.039 --> 00:38:04.320
in this space of, you know, forecast of

00:38:02.519 --> 00:38:06.759
inflation and so on,

00:38:04.320 --> 00:38:08.760
thinks that that that we will need some

00:38:06.760 --> 00:38:11.120
some adjustment through this this part

00:38:08.760 --> 00:38:15.040
as well. Okay.

00:38:11.119 --> 00:38:17.440
My main concern I I think that

00:38:15.039 --> 00:38:19.759
the the Fed the the path the Fed is

00:38:17.440 --> 00:38:21.720
forecasting is feasible,

00:38:19.760 --> 00:38:22.760
but a very narrow path. I mean, it may

00:38:21.719 --> 00:38:24.879
happen.

00:38:22.760 --> 00:38:26.880
And and to me, whether it they're

00:38:24.880 --> 00:38:29.358
successful at not creating a big mess

00:38:26.880 --> 00:38:31.960
here, I mean, bringing unemployment very

00:38:29.358 --> 00:38:34.159
high in order to bring inflation down,

00:38:31.960 --> 00:38:35.880
has a lot to do with whether

00:38:34.159 --> 00:38:37.920
somehow we manage to keep expected

00:38:35.880 --> 00:38:39.960
inflation anchored.

00:38:37.920 --> 00:38:41.599
And there there was some evidence, I

00:38:39.960 --> 00:38:42.840
think I said that a few lectures ago,

00:38:41.599 --> 00:38:45.519
there was some evidence that in the

00:38:42.840 --> 00:38:45.519
summer of

00:38:45.800 --> 00:38:50.720
uh summer of 2022, I'm from the southern

00:38:49.199 --> 00:38:52.559
hemisphere, so I get always confused

00:38:50.719 --> 00:38:55.279
with summers and and so on.

00:38:52.559 --> 00:38:57.400
So, the in in the summer of 2022, US

00:38:55.280 --> 00:38:59.200
summer of 2022, inflation was becoming

00:38:57.400 --> 00:39:01.200
very unanchored. This guy

00:38:59.199 --> 00:39:04.439
one year expected inflation was creeping

00:39:01.199 --> 00:39:07.159
up to 6% and that was very scary. Okay?

00:39:04.440 --> 00:39:10.200
Because think what happened. If if if

00:39:07.159 --> 00:39:12.039
you get expected inflation at 6%,

00:39:10.199 --> 00:39:13.599
then it's not enough to bring

00:39:12.039 --> 00:39:15.358
unemployment to the natural rate of

00:39:13.599 --> 00:39:17.400
unemployment to get inflation back to

00:39:15.358 --> 00:39:19.519
the 2% we like because you need to bring

00:39:17.400 --> 00:39:22.320
expected inflation down now. And that

00:39:19.519 --> 00:39:24.199
means you need to sort of bring the

00:39:22.320 --> 00:39:26.160
unemployment rate very very high in

00:39:24.199 --> 00:39:28.119
order to re-anchor expectations. So,

00:39:26.159 --> 00:39:29.559
that's a very scary situation. They were

00:39:28.119 --> 00:39:32.480
very persuasive though at the end of the

00:39:29.559 --> 00:39:33.320
summer with very hawkish speeches and so

00:39:32.480 --> 00:39:35.719
on

00:39:33.320 --> 00:39:37.160
and they managed to re-anchor expected

00:39:35.719 --> 00:39:38.719
inflation. So, expected inflation very

00:39:37.159 --> 00:39:41.519
quickly came down to two two and a half

00:39:38.719 --> 00:39:43.199
percent one year out to

00:39:41.519 --> 00:39:44.960
But, now it been picking up again and

00:39:43.199 --> 00:39:47.239
now we are around 3% again, so it's a

00:39:44.960 --> 00:39:49.199
little bit scary for where we are. So,

00:39:47.239 --> 00:39:51.199
to me this is going to be very important

00:39:49.199 --> 00:39:53.679
in that. So,

00:39:51.199 --> 00:39:55.919
if inflation keeps lingering around 6%

00:39:53.679 --> 00:39:57.679
and so on, and eventually the expected

00:39:55.920 --> 00:39:59.519
inflation becomes an anchor, then

00:39:57.679 --> 00:40:02.039
there's almost no way around but to have

00:39:59.519 --> 00:40:04.239
a recession to get out of that.

00:40:02.039 --> 00:40:06.279
If that doesn't happen, if they succeed

00:40:04.239 --> 00:40:07.599
convincing a ton of people that that,

00:40:06.280 --> 00:40:09.519
you know, they're very serious about

00:40:07.599 --> 00:40:11.719
about this stuff and they they re-anchor

00:40:09.519 --> 00:40:14.239
expectation expected inflation, then we

00:40:11.719 --> 00:40:15.599
don't need to create a large recession.

00:40:14.239 --> 00:40:17.599
Still they may create it, cause it

00:40:15.599 --> 00:40:20.599
because, you know, accidents happen, but

00:40:17.599 --> 00:40:22.799
but but but they don't need to.

00:40:20.599 --> 00:40:24.119
But they will need to if this guy gets

00:40:22.800 --> 00:40:26.320
an anchor.

00:40:24.119 --> 00:40:28.359
Actually, maybe I can use even this

00:40:26.320 --> 00:40:30.400
expression here

00:40:28.360 --> 00:40:31.920
to explain what I'm trying to say and I

00:40:30.400 --> 00:40:33.880
realize that this is again, this is

00:40:31.920 --> 00:40:35.360
material really for

00:40:33.880 --> 00:40:36.920
for the next lecture.

00:40:35.360 --> 00:40:38.480
What I'm trying to say is that if they

00:40:36.920 --> 00:40:42.840
manage

00:40:38.480 --> 00:40:43.960
to keep this theta very close to zero,

00:40:42.840 --> 00:40:46.960
okay?

00:40:43.960 --> 00:40:50.280
Then, in order to bring inflation back

00:40:46.960 --> 00:40:52.280
to their target of pi bar, 2% or so,

00:40:50.280 --> 00:40:53.680
all that they really need to do is to

00:40:52.280 --> 00:40:55.200
sort of bring unemployment to the

00:40:53.679 --> 00:40:57.399
natural rate of unemployment. So, they

00:40:55.199 --> 00:40:59.119
only need to really

00:40:57.400 --> 00:41:01.280
uh

00:40:59.119 --> 00:41:03.719
fix this gap.

00:41:01.280 --> 00:41:05.760
Okay? They need to raise unemployment so

00:41:03.719 --> 00:41:06.959
so it closes that gap. But it's a small

00:41:05.760 --> 00:41:09.480
change.

00:41:06.960 --> 00:41:12.599
That's if they succeed keeping expected

00:41:09.480 --> 00:41:15.679
inflation at around 2%.

00:41:12.599 --> 00:41:15.679
If they don't,

00:41:16.559 --> 00:41:19.679
say suppose that that

00:41:18.719 --> 00:41:24.119
that

00:41:19.679 --> 00:41:25.759
theta becomes very far from from zero,

00:41:24.119 --> 00:41:28.319
then we have a problem because then

00:41:25.760 --> 00:41:30.760
expected inflation is above the target,

00:41:28.320 --> 00:41:32.240
no? Because we have 6%, so suppose theta

00:41:30.760 --> 00:41:33.760
is equal to one, we have 6%, then

00:41:32.239 --> 00:41:35.679
expected inflation

00:41:33.760 --> 00:41:38.320
is 6%.

00:41:35.679 --> 00:41:40.039
That means that if you if your expected

00:41:38.320 --> 00:41:41.280
inflation

00:41:40.039 --> 00:41:43.000
is

00:41:41.280 --> 00:41:44.560
6%,

00:41:43.000 --> 00:41:47.000
then in order to bring bring the

00:41:44.559 --> 00:41:48.320
inflation if you bring unemployment just

00:41:47.000 --> 00:41:50.159
to the natural rate of unemployment, so

00:41:48.320 --> 00:41:52.120
the red line to the blue line, you

00:41:50.159 --> 00:41:53.559
haven't made a lot of progress. All that

00:41:52.119 --> 00:41:55.559
you have done is

00:41:53.559 --> 00:41:58.920
you have brought down inflation

00:41:55.559 --> 00:42:00.599
to 6%, which is expected inflation.

00:41:58.920 --> 00:42:03.200
So, if you are have expected inflation

00:42:00.599 --> 00:42:05.279
of 6%, you need to bring unemployment

00:42:03.199 --> 00:42:07.519
much higher than the natural rate of

00:42:05.280 --> 00:42:10.320
unemployment in order to bring inflation

00:42:07.519 --> 00:42:11.759
back to the target of 2%.

00:42:10.320 --> 00:42:13.039
That's the reason I say

00:42:11.760 --> 00:42:15.440
to me

00:42:13.039 --> 00:42:18.320
the fight will be

00:42:15.440 --> 00:42:21.159
the battle will be won or lost

00:42:18.320 --> 00:42:23.320
on that term there.

00:42:21.159 --> 00:42:23.319
Yep.

00:42:23.360 --> 00:42:26.519
How much

00:42:24.639 --> 00:42:28.920
of this current like inflationary

00:42:26.519 --> 00:42:30.440
pressure is caused by unemployment? How

00:42:28.920 --> 00:42:31.639
much of it is caused on the supply side?

00:42:30.440 --> 00:42:33.880
Cuz it feels like a lot of this stuff

00:42:31.639 --> 00:42:35.400
like CPI going up, energy prices going

00:42:33.880 --> 00:42:36.760
up, it's like how much can the Fed keep

00:42:35.400 --> 00:42:40.079
control of something like Well, it

00:42:36.760 --> 00:42:41.560
varies a little from different

00:42:40.079 --> 00:42:42.719
This is around the world, but but in the

00:42:41.559 --> 00:42:44.440
US,

00:42:42.719 --> 00:42:47.159
uh for a while a big component of

00:42:44.440 --> 00:42:49.360
inflation was all that stuff.

00:42:47.159 --> 00:42:51.000
Uh you know, bottlenecks in the ports

00:42:49.360 --> 00:42:52.960
and and stuff like that.

00:42:51.000 --> 00:42:54.599
That's almost all gone.

00:42:52.960 --> 00:42:56.039
There's very little of that left. So,

00:42:54.599 --> 00:42:57.880
now is

00:42:56.039 --> 00:42:59.000
is aggregate demand. People feel very

00:42:57.880 --> 00:43:00.599
rich

00:42:59.000 --> 00:43:01.960
for a variety of reasons, they're

00:43:00.599 --> 00:43:04.039
spending a lot and that's the reason

00:43:01.960 --> 00:43:05.840
unemployment is very low.

00:43:04.039 --> 00:43:07.880
It's not unemployment per se, it's just

00:43:05.840 --> 00:43:09.039
the aggregate demand is very high.

00:43:07.880 --> 00:43:10.760
You know?

00:43:09.039 --> 00:43:12.199
Uh and that translates into very low

00:43:10.760 --> 00:43:13.800
unemployment and that feeds into

00:43:12.199 --> 00:43:15.559
inflation this way

00:43:13.800 --> 00:43:17.080
through wages and so on.

00:43:15.559 --> 00:43:18.880
But

00:43:17.079 --> 00:43:20.759
in the US, the component of aggregate

00:43:18.880 --> 00:43:22.880
demand is much larger than in Europe. In

00:43:20.760 --> 00:43:25.400
Europe, those supply side factors are

00:43:22.880 --> 00:43:29.440
much more important. So,

00:43:25.400 --> 00:43:29.440
you know, around the

00:43:29.599 --> 00:43:36.000
Yeah, the summer of 2022, you could say

00:43:33.199 --> 00:43:37.799
both both Europe and the US had about

00:43:36.000 --> 00:43:41.159
the same amount of excess inflation.

00:43:37.800 --> 00:43:44.039
They were all with about 10% inflation.

00:43:41.159 --> 00:43:45.199
But in the US was 2/3 excess aggregate

00:43:44.039 --> 00:43:48.039
demand,

00:43:45.199 --> 00:43:49.399
while in Europe was 2/3 problems on the

00:43:48.039 --> 00:43:51.079
supply side, especially because of the

00:43:49.400 --> 00:43:51.720
war and stuff like that.

00:43:51.079 --> 00:43:52.840
Okay?

00:43:51.719 --> 00:43:54.959
So,

00:43:52.840 --> 00:43:56.519
so but it for the US today is mostly an

00:43:54.960 --> 00:43:57.960
aggregate demand problem. We're not

00:43:56.519 --> 00:43:59.480
going to get a lot of

00:43:57.960 --> 00:44:01.039
Obviously, if the war stops, that's

00:43:59.480 --> 00:44:02.920
going to help,

00:44:01.039 --> 00:44:04.239
but it's not going to be enough. We we

00:44:02.920 --> 00:44:05.920
we need to

00:44:04.239 --> 00:44:07.799
just the economy is too hot. It's too

00:44:05.920 --> 00:44:09.119
much aggregate demand out there.

00:44:07.800 --> 00:44:11.480
That's the that's the fundamental

00:44:09.119 --> 00:44:13.799
problem. Yeah.

00:44:11.480 --> 00:44:16.519
Can you explain again why an increase in

00:44:13.800 --> 00:44:19.519
Z would increase the natural rate of

00:44:16.519 --> 00:44:20.880
unemployment? An increase in Z? Yeah.

00:44:19.519 --> 00:44:21.960
So,

00:44:20.880 --> 00:44:23.599
uh

00:44:21.960 --> 00:44:25.920
um

00:44:23.599 --> 00:44:27.960
for that the basis the previous slide

00:44:25.920 --> 00:44:30.280
diagram, but remember what Z does.

00:44:27.960 --> 00:44:33.440
Actually, let me go to

00:44:30.280 --> 00:44:33.440
this equation here.

00:44:34.159 --> 00:44:39.079
So, we can figure out in this in this

00:44:36.039 --> 00:44:40.400
two equations here. If Z goes up, that

00:44:39.079 --> 00:44:42.880
means for any given level of

00:44:40.400 --> 00:44:45.119
unemployment

00:44:42.880 --> 00:44:49.960
and expected inflation,

00:44:45.119 --> 00:44:49.960
wages go up. Workers demand higher wage.

00:44:50.000 --> 00:44:55.119
But

00:44:51.559 --> 00:44:56.199
remember that that the firms

00:44:55.119 --> 00:44:58.000
uh

00:44:56.199 --> 00:44:58.919
So, so let me let me let me we're

00:44:58.000 --> 00:45:00.760
talking about the natural rate of

00:44:58.920 --> 00:45:02.480
unemployment, so let me replace this PE

00:45:00.760 --> 00:45:03.359
for P first of all.

00:45:02.480 --> 00:45:05.320
Okay?

00:45:03.358 --> 00:45:09.880
So, I'm going to divide

00:45:05.320 --> 00:45:12.240
W by P both sides. So, I get

00:45:09.880 --> 00:45:14.800
if if Z goes up, the workers want a

00:45:12.239 --> 00:45:17.439
higher real wage.

00:45:14.800 --> 00:45:20.880
No? If because

00:45:17.440 --> 00:45:23.200
if Z goes up, then W over P, I'm

00:45:20.880 --> 00:45:25.680
dividing by P both sides, goes up.

00:45:23.199 --> 00:45:27.439
Workers demand a higher wage.

00:45:25.679 --> 00:45:29.879
But the firms, from here you can see

00:45:27.440 --> 00:45:32.400
that I can divide by P both sides, W

00:45:29.880 --> 00:45:35.200
over P that the firms offer is equal to

00:45:32.400 --> 00:45:37.760
1 over 1 + M.

00:45:35.199 --> 00:45:40.159
Okay? So, the the firms are not going to

00:45:37.760 --> 00:45:42.440
offer a higher real wage. The workers

00:45:40.159 --> 00:45:43.480
want a higher real wage.

00:45:42.440 --> 00:45:45.400
The only thing that can restore

00:45:43.480 --> 00:45:47.519
equilibrium that the workers end up

00:45:45.400 --> 00:45:49.358
demanding the same real wage as the

00:45:47.519 --> 00:45:51.719
firms are willing to pay

00:45:49.358 --> 00:45:53.920
is that somehow the hands of the worker

00:45:51.719 --> 00:45:56.480
gets weakened. And the only variable

00:45:53.920 --> 00:45:58.559
here that can weaken their hand is a

00:45:56.480 --> 00:45:59.358
higher unemployment.

00:45:58.559 --> 00:46:00.799
Okay?

00:45:59.358 --> 00:46:03.719
So,

00:46:00.800 --> 00:46:03.720
let me put it all in

00:46:05.679 --> 00:46:10.358
So, at the natural rate,

00:46:07.599 --> 00:46:13.319
I know that PE is equal to P.

00:46:10.358 --> 00:46:16.239
So, that means the wage setting equation

00:46:13.320 --> 00:46:19.120
the wage setting equation implies

00:46:16.239 --> 00:46:22.479
W over P

00:46:19.119 --> 00:46:22.480
equal F U Z.

00:46:22.920 --> 00:46:28.440
Okay?

00:46:24.800 --> 00:46:28.440
From the price setting equation,

00:46:28.480 --> 00:46:32.079
I have that

00:46:29.840 --> 00:46:35.160
W over P

00:46:32.079 --> 00:46:37.000
is equal to 1 over 1 + M.

00:46:35.159 --> 00:46:38.399
So, in this very simple model, this is

00:46:37.000 --> 00:46:40.440
given.

00:46:38.400 --> 00:46:42.400
If this guy goes up,

00:46:40.440 --> 00:46:43.800
these guys want a higher real wage, but

00:46:42.400 --> 00:46:45.240
that cannot happen because that would be

00:46:43.800 --> 00:46:48.039
inconsistent with the price setting, so

00:46:45.239 --> 00:46:49.759
you need to bring down this guy down.

00:46:48.039 --> 00:46:52.039
The only thing that can bring it down is

00:46:49.760 --> 00:46:54.080
for unemployment to go up.

00:46:52.039 --> 00:46:56.119
And that's at P, we call that the

00:46:54.079 --> 00:46:58.199
natural rate of unemployment.

00:46:56.119 --> 00:46:58.199
Okay.

00:46:59.880 --> 00:47:03.800
Yeah.

00:47:01.320 --> 00:47:07.720
So, like last lecture we talked about

00:47:03.800 --> 00:47:10.640
the labor force participation rate. Um

00:47:07.719 --> 00:47:12.839
is there like any reason to try and like

00:47:10.639 --> 00:47:15.679
increase that to increase Oh,

00:47:12.840 --> 00:47:15.680
fantastic. Yes.

00:47:16.880 --> 00:47:21.160
Well, I mean

00:47:19.039 --> 00:47:22.119
there are sort of negative policies as

00:47:21.159 --> 00:47:24.519
well.

00:47:22.119 --> 00:47:25.960
You know, Z reduction in a sense does

00:47:24.519 --> 00:47:28.079
that because

00:47:25.960 --> 00:47:30.480
the the was a emergency unemployment

00:47:28.079 --> 00:47:32.920
benefits and emergency

00:47:30.480 --> 00:47:34.599
income supplements and so on as a result

00:47:32.920 --> 00:47:36.480
of the pandemic that are disappearing

00:47:34.599 --> 00:47:39.400
slowly. And that's very naturally so

00:47:36.480 --> 00:47:41.159
it's it's going to bring

00:47:39.400 --> 00:47:43.519
uh participation back up and it is

00:47:41.159 --> 00:47:46.079
beginning to pick up. So,

00:47:43.519 --> 00:47:48.079
so yeah, you need to incentivize return

00:47:46.079 --> 00:47:48.920
to work. And now there are some people

00:47:48.079 --> 00:47:50.880
that

00:47:48.920 --> 00:47:52.480
there's nothing that

00:47:50.880 --> 00:47:54.320
they've retired essentially or, you

00:47:52.480 --> 00:47:57.199
know, they have health problems and they

00:47:54.320 --> 00:47:58.480
they just cannot return. We lost that.

00:47:57.199 --> 00:48:00.399
And the other margin which is very

00:47:58.480 --> 00:48:01.840
important is immigration. So, that's a

00:48:00.400 --> 00:48:03.440
big issue

00:48:01.840 --> 00:48:05.440
because immigration obviously that we

00:48:03.440 --> 00:48:07.119
lost I think in the US, I'm not a labor

00:48:05.440 --> 00:48:09.280
economist, but we lost

00:48:07.119 --> 00:48:11.559
I think a flow of the order of the order

00:48:09.280 --> 00:48:13.519
of 500,000 people a year

00:48:11.559 --> 00:48:15.920
during COVID.

00:48:13.519 --> 00:48:17.320
And and and that's that's a big chunk of

00:48:15.920 --> 00:48:19.119
the decline in

00:48:17.320 --> 00:48:21.640
in the labor No, what you need is more

00:48:19.119 --> 00:48:23.279
employment. That's going to that puts

00:48:21.639 --> 00:48:25.159
downward pressure on wages for the same

00:48:23.280 --> 00:48:28.240
amount of aggregate demand.

00:48:25.159 --> 00:48:29.440
And that's what you need, but but

00:48:28.239 --> 00:48:31.000
Yeah, we're taking that's a very good

00:48:29.440 --> 00:48:33.559
point. We're taking all that as given

00:48:31.000 --> 00:48:36.480
here. Remember, we're fixing all that,

00:48:33.559 --> 00:48:38.119
but but if you don't, then then you

00:48:36.480 --> 00:48:40.679
other terms will start appearing in this

00:48:38.119 --> 00:48:42.480
expression and so on.

00:48:40.679 --> 00:48:43.480
Good.

00:48:42.480 --> 00:48:45.880
Obviously, I'm not going to start the

00:48:43.480 --> 00:48:48.920
review. We have only 1 minute, but so in

00:48:45.880 --> 00:48:52.480
the next lecture I I'll just review

00:48:48.920 --> 00:48:52.480
uh the material for the quiz.