WEBVTT 00:00:00.000 --> 00:00:01.976 align:middle line:90% [SQUEAKING] 00:00:01.976 --> 00:00:03.952 align:middle line:90% [RUSTLING] 00:00:03.952 --> 00:00:06.422 align:middle line:90% [CLICKING] 00:00:06.422 --> 00:00:17.790 align:middle line:90% 00:00:17.790 --> 00:00:19.680 align:middle line:84% RICARDO CABALLERO: So after doing IS-LM 00:00:19.680 --> 00:00:21.990 align:middle line:84% in the first part of the course and where 00:00:21.990 --> 00:00:25.170 align:middle line:84% we took prices completely sticky and output was fully 00:00:25.170 --> 00:00:27.840 align:middle line:84% determined by aggregate demand, we 00:00:27.840 --> 00:00:31.600 align:middle line:84% said, well, that dominates in the very, very short run. 00:00:31.600 --> 00:00:35.893 align:middle line:84% But over time, at some point the supply side starts showing up. 00:00:35.893 --> 00:00:36.810 align:middle line:90% There are constraints. 00:00:36.810 --> 00:00:39.610 align:middle line:84% The labor market gets very tight, and so on. 00:00:39.610 --> 00:00:45.130 align:middle line:84% And so we added a block that started from wage determination. 00:00:45.130 --> 00:00:48.600 align:middle line:84% And then we look at the impact of wages on prices. 00:00:48.600 --> 00:00:51.600 align:middle line:84% And then we related inflation rate. 00:00:51.600 --> 00:00:53.430 align:middle line:84% You start to relate the inflation rate 00:00:53.430 --> 00:00:55.570 align:middle line:90% to economic activity. 00:00:55.570 --> 00:01:00.260 align:middle line:84% So output above or below the potential output, 00:01:00.260 --> 00:01:03.550 align:middle line:84% or the natural level of output and things of that kind. 00:01:03.550 --> 00:01:09.880 align:middle line:84% So remember, the starting point was a wage demand equation, 00:01:09.880 --> 00:01:12.780 align:middle line:84% so what workers demand for a wage. 00:01:12.780 --> 00:01:15.210 align:middle line:84% This period depends on what is the price level 00:01:15.210 --> 00:01:17.123 align:middle line:90% they expect for the period. 00:01:17.123 --> 00:01:18.790 align:middle line:84% Because they set the wage today and they 00:01:18.790 --> 00:01:20.770 align:middle line:84% have to live through the year or two, 00:01:20.770 --> 00:01:25.250 align:middle line:84% whatever is the contracting period, with that nominal wage. 00:01:25.250 --> 00:01:28.420 align:middle line:84% So naturally, if they expect higher price level 00:01:28.420 --> 00:01:31.240 align:middle line:84% in the future, they're going to demand a higher nominal wage 00:01:31.240 --> 00:01:32.120 align:middle line:90% today. 00:01:32.120 --> 00:01:34.630 align:middle line:84% And then we said that's a function that is also 00:01:34.630 --> 00:01:36.820 align:middle line:84% going to be decreasing in the level of unemployment 00:01:36.820 --> 00:01:42.310 align:middle line:84% because obviously that weakens bargaining power for workers 00:01:42.310 --> 00:01:47.200 align:middle line:84% or makes actually becoming unemployed or not 00:01:47.200 --> 00:01:51.430 align:middle line:84% having a job more costly because it's very difficult to exit out 00:01:51.430 --> 00:01:52.630 align:middle line:90% of unemployment. 00:01:52.630 --> 00:01:55.660 align:middle line:84% And then we made normalization, this function, also 00:01:55.660 --> 00:01:57.970 align:middle line:84% an increasing function on this variable z, 00:01:57.970 --> 00:02:00.850 align:middle line:84% which captures a bunch of labor market institutions, 00:02:00.850 --> 00:02:03.740 align:middle line:84% including wage labor bargaining power. 00:02:03.740 --> 00:02:06.100 align:middle line:84% So more bargaining power means that for any given 00:02:06.100 --> 00:02:08.229 align:middle line:84% level of unemployment, workers would 00:02:08.229 --> 00:02:10.150 align:middle line:90% tend to demand a higher wage. 00:02:10.150 --> 00:02:14.290 align:middle line:84% So that's what the z variable was all about. 00:02:14.290 --> 00:02:17.620 align:middle line:84% Then we wanted to go from wages to prices, 00:02:17.620 --> 00:02:20.440 align:middle line:84% because the ultimate goal was to bring inflation 00:02:20.440 --> 00:02:21.560 align:middle line:90% into the picture. 00:02:21.560 --> 00:02:23.800 align:middle line:84% And for that, we had to produce-- 00:02:23.800 --> 00:02:30.260 align:middle line:84% we introduce a production function because in particular, 00:02:30.260 --> 00:02:33.010 align:middle line:84% we made output a function of employment. 00:02:33.010 --> 00:02:36.370 align:middle line:84% And that, very naturally, will connect wage pressure 00:02:36.370 --> 00:02:40.090 align:middle line:84% to price pressure because you need labor to produce output. 00:02:40.090 --> 00:02:43.000 align:middle line:84% So that the labor market is very tight, that means also 00:02:43.000 --> 00:02:45.760 align:middle line:84% there's going to be more expensive to produce output. 00:02:45.760 --> 00:02:48.170 align:middle line:84% And we simplify this production function a lot. 00:02:48.170 --> 00:02:51.470 align:middle line:84% We made it output equal to employment. 00:02:51.470 --> 00:02:55.750 align:middle line:84% And that meant also that one unit of labor, 00:02:55.750 --> 00:02:58.150 align:middle line:84% in order to produce one extra unit of output, 00:02:58.150 --> 00:03:00.520 align:middle line:84% you need one extra unit of labor, 00:03:00.520 --> 00:03:05.960 align:middle line:84% which means you need to pay a wage, one unit of the wage. 00:03:05.960 --> 00:03:09.010 align:middle line:84% And so then we said, suppose that the price setting 00:03:09.010 --> 00:03:11.020 align:middle line:84% from the side of the firms simply 00:03:11.020 --> 00:03:14.293 align:middle line:84% takes this cost, which is the wage, and adds a markup to it 00:03:14.293 --> 00:03:15.710 align:middle line:84% to pay for a bunch of other things 00:03:15.710 --> 00:03:19.100 align:middle line:84% that we haven't introduced in this model. 00:03:19.100 --> 00:03:22.550 align:middle line:84% So the price charged by firms is equal to the wage times 1 00:03:22.550 --> 00:03:26.850 align:middle line:84% plus some positive numbers, 8.2 or something like that, so 1.2. 00:03:26.850 --> 00:03:29.970 align:middle line:84% And we can rewrite this price setting equation 00:03:29.970 --> 00:03:34.530 align:middle line:84% as a wage, a real wage the firms are willing to offer. 00:03:34.530 --> 00:03:36.420 align:middle line:90% And it's just equal to that. 00:03:36.420 --> 00:03:38.850 align:middle line:84% So when the markup goes up, that means 00:03:38.850 --> 00:03:41.160 align:middle line:84% the real wage firms are willing to offer 00:03:41.160 --> 00:03:42.830 align:middle line:90% is lower than otherwise. 00:03:42.830 --> 00:03:45.590 align:middle line:90% 00:03:45.590 --> 00:03:48.020 align:middle line:84% So that took us to the concept of the natural rate 00:03:48.020 --> 00:03:48.800 align:middle line:90% of unemployment. 00:03:48.800 --> 00:03:52.670 align:middle line:84% And what I said, no, there's nothing 00:03:52.670 --> 00:03:55.110 align:middle line:84% natural about the natural rate of unemployment. 00:03:55.110 --> 00:03:57.500 align:middle line:84% It's simply a definition that says 00:03:57.500 --> 00:04:01.790 align:middle line:84% that's unemployment that results when the price, expected price, 00:04:01.790 --> 00:04:03.180 align:middle line:90% is equal to the actual price. 00:04:03.180 --> 00:04:06.890 align:middle line:90% That's all that is. 00:04:06.890 --> 00:04:09.290 align:middle line:84% And when we have that condition, then we 00:04:09.290 --> 00:04:13.880 align:middle line:84% can think of the real wage demanded by workers because I 00:04:13.880 --> 00:04:16.670 align:middle line:84% can replace the expected price for actual price 00:04:16.670 --> 00:04:18.329 align:middle line:90% and divide both sides by price. 00:04:18.329 --> 00:04:22.010 align:middle line:84% So the actual wage demanded by workers 00:04:22.010 --> 00:04:25.550 align:middle line:84% is equal to a function of the natural rate of unemployment. 00:04:25.550 --> 00:04:27.890 align:middle line:84% And I stick the end there precisely 00:04:27.890 --> 00:04:31.590 align:middle line:84% because I replace expected price or P, for no other reason. 00:04:31.590 --> 00:04:35.180 align:middle line:84% But now we have two equations for the real wage, the real wage 00:04:35.180 --> 00:04:39.170 align:middle line:84% that firms are willing to pay and the real wage and workers 00:04:39.170 --> 00:04:40.230 align:middle line:90% need to demand. 00:04:40.230 --> 00:04:42.210 align:middle line:90% And we can make them both equal. 00:04:42.210 --> 00:04:45.150 align:middle line:84% And that determines the natural rate of unemployment. 00:04:45.150 --> 00:04:48.560 align:middle line:84% OK, so remember this from the point of view of the firm. 00:04:48.560 --> 00:04:50.800 align:middle line:84% This is equal to 1 over 1 plus the markup. 00:04:50.800 --> 00:04:54.040 align:middle line:84% The only endogenous variable in the markup is a constant z. 00:04:54.040 --> 00:04:56.880 align:middle line:84% This is also a parameter is exogenous. 00:04:56.880 --> 00:04:59.960 align:middle line:84% And so from here, we can solve the natural rate 00:04:59.960 --> 00:05:02.660 align:middle line:84% of unemployment, 1 over 1 plus m. 00:05:02.660 --> 00:05:04.920 align:middle line:84% And we can solve the natural rate of unemployment. 00:05:04.920 --> 00:05:07.513 align:middle line:84% And if you do the algebra right, you're 00:05:07.513 --> 00:05:08.930 align:middle line:84% going to get to a point like that. 00:05:08.930 --> 00:05:11.840 align:middle line:84% That pins down the natural rate of unemployment. 00:05:11.840 --> 00:05:14.420 align:middle line:84% Again, there is nothing natural about the natural rate 00:05:14.420 --> 00:05:15.240 align:middle line:90% of unemployment. 00:05:15.240 --> 00:05:19.910 align:middle line:84% It depends on a bunch of parameters, which, for example, 00:05:19.910 --> 00:05:21.780 align:middle line:84% it clearly depends on the markup. 00:05:21.780 --> 00:05:25.310 align:middle line:84% It depends on things that we took as constant here as given 00:05:25.310 --> 00:05:27.830 align:middle line:84% here, all the things that were in z. 00:05:27.830 --> 00:05:29.930 align:middle line:90% Those are part of that. 00:05:29.930 --> 00:05:32.210 align:middle line:84% And so then we look at things that change. 00:05:32.210 --> 00:05:34.140 align:middle line:84% And that's just done with equations. 00:05:34.140 --> 00:05:36.080 align:middle line:84% We look at things that change the natural rate 00:05:36.080 --> 00:05:37.200 align:middle line:90% of unemployment. 00:05:37.200 --> 00:05:38.700 align:middle line:90% That's one example. 00:05:38.700 --> 00:05:40.932 align:middle line:84% If bargaining power by workers goes up, 00:05:40.932 --> 00:05:43.640 align:middle line:84% they're going to demand a higher wage at the initial natural rate 00:05:43.640 --> 00:05:44.730 align:middle line:90% of unemployment. 00:05:44.730 --> 00:05:46.820 align:middle line:84% Well, that obviously, that higher wage 00:05:46.820 --> 00:05:49.350 align:middle line:84% is inconsistent with what firms are willing to pay. 00:05:49.350 --> 00:05:51.770 align:middle line:84% The only way equilibrium can be restored 00:05:51.770 --> 00:05:54.110 align:middle line:84% in this model, that's the medium run equilibrium, 00:05:54.110 --> 00:05:56.120 align:middle line:84% is for the natural rate of unemployment 00:05:56.120 --> 00:05:57.540 align:middle line:90% to rise to UN prime. 00:05:57.540 --> 00:06:00.420 align:middle line:90% 00:06:00.420 --> 00:06:01.190 align:middle line:90% There you have it. 00:06:01.190 --> 00:06:02.325 align:middle line:90% Nothing natural. 00:06:02.325 --> 00:06:04.200 align:middle line:84% Natural rate of unemployment is not constant. 00:06:04.200 --> 00:06:05.850 align:middle line:84% It depends on institutional parameters, 00:06:05.850 --> 00:06:08.490 align:middle line:90% such as bargaining power. 00:06:08.490 --> 00:06:11.430 align:middle line:84% Another example is mark ups. It depends on mark ups, 00:06:11.430 --> 00:06:13.830 align:middle line:84% as well, the degree of competition, if you will, 00:06:13.830 --> 00:06:15.100 align:middle line:90% in the goods market. 00:06:15.100 --> 00:06:18.840 align:middle line:84% If we are in some equilibrium like this one and now 00:06:18.840 --> 00:06:21.930 align:middle line:84% suddenly firms, for whatever reason, 00:06:21.930 --> 00:06:26.730 align:middle line:84% choose or need to charge a higher markup, that 00:06:26.730 --> 00:06:29.850 align:middle line:84% means that at this level of unemployment, 00:06:29.850 --> 00:06:32.730 align:middle line:84% the wage that workers would demand is higher than the wage 00:06:32.730 --> 00:06:35.185 align:middle line:84% that firms are willing to pay, the real wage. 00:06:35.185 --> 00:06:37.560 align:middle line:84% And the only thing that can clear the market in this case 00:06:37.560 --> 00:06:41.220 align:middle line:84% here, in the medium run, is for the natural rate of unemployment 00:06:41.220 --> 00:06:42.360 align:middle line:90% to rise. 00:06:42.360 --> 00:06:44.730 align:middle line:84% So here, we go two experiments where 00:06:44.730 --> 00:06:48.210 align:middle line:84% we move some parameter-- in one, the bargaining power of workers 00:06:48.210 --> 00:06:51.210 align:middle line:84% and the other one the markup of the firms. 00:06:51.210 --> 00:06:54.880 align:middle line:84% And both increase the natural rate of unemployment. 00:06:54.880 --> 00:06:57.390 align:middle line:90% Good. 00:06:57.390 --> 00:07:01.020 align:middle line:84% The next step was to look at things 00:07:01.020 --> 00:07:03.260 align:middle line:84% that happen outside the natural rate of unemployment, 00:07:03.260 --> 00:07:07.290 align:middle line:84% in particular, what happens to prices there. 00:07:07.290 --> 00:07:09.870 align:middle line:90% So what we did is we took the-- 00:07:09.870 --> 00:07:14.798 align:middle line:84% we went back to the model with the expected price here. 00:07:14.798 --> 00:07:16.590 align:middle line:84% That means that unemployment that comes out 00:07:16.590 --> 00:07:18.540 align:middle line:84% from this equilibrium is not going 00:07:18.540 --> 00:07:21.010 align:middle line:84% to be necessarily the natural rate of unemployment. 00:07:21.010 --> 00:07:25.320 align:middle line:84% That will be the case only if P happens to be equal to P. 00:07:25.320 --> 00:07:28.500 align:middle line:84% Then we simplify this function, F, here for something linear, 00:07:28.500 --> 00:07:29.520 align:middle line:90% like this. 00:07:29.520 --> 00:07:31.540 align:middle line:84% Very simple, but again, decreasing 00:07:31.540 --> 00:07:37.150 align:middle line:84% in unemployment, increasing in this institutional parameter, z. 00:07:37.150 --> 00:07:40.750 align:middle line:84% We replace this wage here from this expression 00:07:40.750 --> 00:07:42.380 align:middle line:90% here and rearrange. 00:07:42.380 --> 00:07:45.550 align:middle line:90% So we got this here. 00:07:45.550 --> 00:07:48.370 align:middle line:84% And the next step was just to go from here 00:07:48.370 --> 00:07:49.820 align:middle line:90% to the rate of inflation. 00:07:49.820 --> 00:07:53.200 align:middle line:84% And we did it through several steps and approximations. 00:07:53.200 --> 00:07:57.350 align:middle line:84% And we ended up with what is known as the Phillips curve. 00:07:57.350 --> 00:08:00.730 align:middle line:84% So this says that inflation is increasing and expected 00:08:00.730 --> 00:08:04.510 align:middle line:84% inflation on this institutional parameters. 00:08:04.510 --> 00:08:10.090 align:middle line:84% As the markups go up, that will tend to increase inflation. 00:08:10.090 --> 00:08:14.030 align:middle line:84% If bargaining power by workers go up, then that is the same. 00:08:14.030 --> 00:08:16.330 align:middle line:84% But most importantly, it's negatively 00:08:16.330 --> 00:08:17.720 align:middle line:90% related to unemployment. 00:08:17.720 --> 00:08:20.920 align:middle line:84% And that's the reason that today, nowadays, you 00:08:20.920 --> 00:08:24.070 align:middle line:84% know, there's lots of discussion about the tightness in the labor 00:08:24.070 --> 00:08:26.995 align:middle line:84% market and whether that's really necessary. 00:08:26.995 --> 00:08:29.620 align:middle line:84% Do we need to cause a recession, a situation where unemployment 00:08:29.620 --> 00:08:32.530 align:middle line:84% goes up a lot, in order to really finally bring down 00:08:32.530 --> 00:08:33.429 align:middle line:90% inflation? 00:08:33.429 --> 00:08:36.065 align:middle line:90% Yeah, there was a question. 00:08:36.065 --> 00:08:36.940 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:08:36.940 --> 00:08:38.260 align:middle line:90% RICARDO CABALLERO: Yeah. 00:08:38.260 --> 00:08:40.630 align:middle line:84% Remember that I made up this function. 00:08:40.630 --> 00:08:43.929 align:middle line:84% We said this function is decreasing unemployment. 00:08:43.929 --> 00:08:48.690 align:middle line:84% I just replace that function for that. 00:08:48.690 --> 00:08:55.750 align:middle line:84% So it's the sensitivity of wage demand by workers 00:08:55.750 --> 00:08:58.270 align:middle line:90% to their employment rate. 00:08:58.270 --> 00:09:00.910 align:middle line:84% Alpha is very high, it means that wage demand 00:09:00.910 --> 00:09:04.363 align:middle line:84% is very sensitive, very responsive to unemployment. 00:09:04.363 --> 00:09:06.863 align:middle line:84% AUDIENCE: What would be the situation for an expected price? 00:09:06.863 --> 00:09:09.196 align:middle line:84% Like, could you connect that back to like, I don't know, 00:09:09.196 --> 00:09:11.290 align:middle line:84% some sort of a commodity or something? 00:09:11.290 --> 00:09:13.730 align:middle line:84% RICARDO CABALLERO: So what is the intuition for this? 00:09:13.730 --> 00:09:16.770 align:middle line:84% AUDIENCE: Yeah, just like a price feels tactile. 00:09:16.770 --> 00:09:18.810 align:middle line:90% But an expected price, I don't-- 00:09:18.810 --> 00:09:22.560 align:middle line:84% RICARDO CABALLERO: Well, I mean, imagine that workers and firms 00:09:22.560 --> 00:09:25.510 align:middle line:84% bargain for a wage that will live through the year. 00:09:25.510 --> 00:09:29.410 align:middle line:84% You're bargaining for the wage, nominal wage today. 00:09:29.410 --> 00:09:30.580 align:middle line:90% You don't set a real wage. 00:09:30.580 --> 00:09:34.020 align:middle line:84% You set the nominal wage, say $100, whatever. 00:09:34.020 --> 00:09:36.510 align:middle line:84% Well, the wage demand will depend a lot 00:09:36.510 --> 00:09:39.670 align:middle line:84% on what they expect inflation to be during this period. 00:09:39.670 --> 00:09:42.360 align:middle line:84% If I expect inflation to be 10%, you're 00:09:42.360 --> 00:09:44.345 align:middle line:84% very likely to demand a higher nominal wage 00:09:44.345 --> 00:09:46.720 align:middle line:84% because you have to live, on average, with higher prices. 00:09:46.720 --> 00:09:49.000 align:middle line:84% So that's the role of that, is the price. 00:09:49.000 --> 00:09:50.670 align:middle line:90% I mean, I would prefer-- 00:09:50.670 --> 00:09:52.980 align:middle line:84% and there are countries where that's done-- 00:09:52.980 --> 00:09:55.058 align:middle line:84% to set my wage in real terms so I 00:09:55.058 --> 00:09:56.350 align:middle line:90% don't need to worry about that. 00:09:56.350 --> 00:09:59.430 align:middle line:84% But in practice, in economies with low inflation like the US, 00:09:59.430 --> 00:10:00.250 align:middle line:90% you don't do that. 00:10:00.250 --> 00:10:03.480 align:middle line:84% You get a nominal wage and you have to live for the year 00:10:03.480 --> 00:10:06.490 align:middle line:84% or until the next negotiation for your wage contract 00:10:06.490 --> 00:10:09.548 align:middle line:90% with that level of wages. 00:10:09.548 --> 00:10:11.840 align:middle line:84% AUDIENCE: [INAUDIBLE] be dependent on time, [INAUDIBLE] 00:10:11.840 --> 00:10:15.770 align:middle line:84% with the interest rate or with the inflation rate, 00:10:15.770 --> 00:10:19.700 align:middle line:84% whereas I guess the regular price is defined by the wage is 00:10:19.700 --> 00:10:20.810 align:middle line:90% dependent on the market? 00:10:20.810 --> 00:10:22.685 align:middle line:84% RICARDO CABALLERO: No, they're both the same. 00:10:22.685 --> 00:10:28.690 align:middle line:84% But the only thing is that this price here is not the current. 00:10:28.690 --> 00:10:31.690 align:middle line:84% It's what you really expect the price to be during the year. 00:10:31.690 --> 00:10:34.210 align:middle line:84% If this is here just because at the moment in which you 00:10:34.210 --> 00:10:36.580 align:middle line:84% set the wage, you don't know the price 00:10:36.580 --> 00:10:38.560 align:middle line:84% that you're going to face as a worker. 00:10:38.560 --> 00:10:40.270 align:middle line:90% But it's the price. 00:10:40.270 --> 00:10:43.450 align:middle line:84% So you don't know the price that you're going to actually face. 00:10:43.450 --> 00:10:46.390 align:middle line:84% So the best you can do is calculate, well, 00:10:46.390 --> 00:10:48.320 align:middle line:84% I think inflation is going to be 10%. 00:10:48.320 --> 00:10:50.770 align:middle line:84% So give me what I would have had in mind 00:10:50.770 --> 00:10:53.298 align:middle line:84% with inflation equal to 0 plus 5%, 00:10:53.298 --> 00:10:54.590 align:middle line:90% so on average, I'm about right. 00:10:54.590 --> 00:10:56.830 align:middle line:90% That's the logic. 00:10:56.830 --> 00:10:59.878 align:middle line:84% But this expected price is meant to be your best 00:10:59.878 --> 00:11:02.170 align:middle line:84% proxy you have at the moment in which you're bargaining 00:11:02.170 --> 00:11:04.480 align:middle line:84% for your wage for what the actual price will 00:11:04.480 --> 00:11:07.890 align:middle line:84% be during the life of that particular wage. 00:11:07.890 --> 00:11:16.650 align:middle line:90% 00:11:16.650 --> 00:11:21.410 align:middle line:84% So we end up with that Phillips curve here. 00:11:21.410 --> 00:11:23.730 align:middle line:84% Importantly, this is a decreasing function 00:11:23.730 --> 00:11:26.050 align:middle line:90% of unemployment. 00:11:26.050 --> 00:11:29.922 align:middle line:84% And then we made different assumptions about expectations. 00:11:29.922 --> 00:11:31.380 align:middle line:84% If expected inflation, for example, 00:11:31.380 --> 00:11:34.860 align:middle line:84% is a constant, that's when we say expected inflation is very 00:11:34.860 --> 00:11:38.220 align:middle line:84% well anchored, then you get a Phillips curve 00:11:38.220 --> 00:11:41.790 align:middle line:84% that looks like this, in which inflation has a constant here 00:11:41.790 --> 00:11:43.990 align:middle line:84% and a decreasing on the rate of unemployment. 00:11:43.990 --> 00:11:48.450 align:middle line:84% And during the '60s, that relationship sort of held fairly 00:11:48.450 --> 00:11:49.240 align:middle line:90% well. 00:11:49.240 --> 00:11:51.040 align:middle line:84% It was a downward sloping relationship. 00:11:51.040 --> 00:11:54.510 align:middle line:84% It got to be steeper and steeper as we moved into higher 00:11:54.510 --> 00:11:55.780 align:middle line:90% and higher inflation levels. 00:11:55.780 --> 00:11:58.470 align:middle line:84% And then I said, but in the '70s, the whole thing broke 00:11:58.470 --> 00:11:58.980 align:middle line:90% loose. 00:11:58.980 --> 00:12:02.620 align:middle line:84% There's nothing like a downward sloping curve here. 00:12:02.620 --> 00:12:04.210 align:middle line:90% That happened for two reasons. 00:12:04.210 --> 00:12:05.680 align:middle line:84% There were some cost push shocks. 00:12:05.680 --> 00:12:07.480 align:middle line:84% You can think of lots of shocks to them. 00:12:07.480 --> 00:12:11.970 align:middle line:84% But more interestingly, expected inflation became an anchor. 00:12:11.970 --> 00:12:15.690 align:middle line:84% And then we changed the expected inflation model 00:12:15.690 --> 00:12:17.260 align:middle line:84% for rather than being a constant, 00:12:17.260 --> 00:12:19.610 align:middle line:84% being some weighted average like this. 00:12:19.610 --> 00:12:25.000 align:middle line:84% And we said, look, during the '70s essentially that theta was 00:12:25.000 --> 00:12:26.030 align:middle line:90% equal to 1. 00:12:26.030 --> 00:12:29.620 align:middle line:84% So expected inflation was really-- 00:12:29.620 --> 00:12:31.180 align:middle line:84% whatever was inflation last year, 00:12:31.180 --> 00:12:32.920 align:middle line:84% people expected that level of inflation 00:12:32.920 --> 00:12:35.470 align:middle line:84% to stay in the next year, rather than going back 00:12:35.470 --> 00:12:38.260 align:middle line:84% to that whatever was the constant or inflation target 00:12:38.260 --> 00:12:41.200 align:middle line:90% or historical constant pi. 00:12:41.200 --> 00:12:43.948 align:middle line:84% And that meant that during that period, 00:12:43.948 --> 00:12:45.490 align:middle line:84% really the Phillips curve looked more 00:12:45.490 --> 00:12:49.180 align:middle line:84% like a relationship of the change in the inflation rate 00:12:49.180 --> 00:12:51.170 align:middle line:84% as a decreasing function of unemployment. 00:12:51.170 --> 00:12:54.040 align:middle line:84% So that means that when you increase unemployment here, 00:12:54.040 --> 00:12:56.920 align:middle line:84% you reduce the rate at which unemployment, inflation is 00:12:56.920 --> 00:12:57.650 align:middle line:90% rising. 00:12:57.650 --> 00:13:01.690 align:middle line:84% That's the goal of the situation in a case 00:13:01.690 --> 00:13:06.050 align:middle line:84% in which expected inflation is an anchor. 00:13:06.050 --> 00:13:09.185 align:middle line:84% And the last step we had there is we replace-- 00:13:09.185 --> 00:13:10.810 align:middle line:84% we noticed, we said, well, what happens 00:13:10.810 --> 00:13:13.630 align:middle line:84% if we stick in here the natural rate of unemployment? 00:13:13.630 --> 00:13:16.090 align:middle line:84% Then that will give us-- that will happen only 00:13:16.090 --> 00:13:18.200 align:middle line:84% when expected price is equal to actual prices. 00:13:18.200 --> 00:13:19.840 align:middle line:84% So that means that when inflation 00:13:19.840 --> 00:13:22.360 align:middle line:84% is equal to expected inflation, from here we 00:13:22.360 --> 00:13:24.220 align:middle line:84% can solve the natural rate of unemployment 00:13:24.220 --> 00:13:27.080 align:middle line:84% as a function of these structural parameters. 00:13:27.080 --> 00:13:30.460 align:middle line:84% And once we have that, we could go back to our Phillips curve 00:13:30.460 --> 00:13:32.960 align:middle line:90% and rewrite it in this way. 00:13:32.960 --> 00:13:36.190 align:middle line:84% So you can think of the Phillips curve in this way. 00:13:36.190 --> 00:13:39.670 align:middle line:84% And this is the way we typically write it down, 00:13:39.670 --> 00:13:44.470 align:middle line:84% in which it says inflation is decreasing in the unemployment 00:13:44.470 --> 00:13:44.970 align:middle line:90% gap. 00:13:44.970 --> 00:13:48.850 align:middle line:90% 00:13:48.850 --> 00:13:51.032 align:middle line:84% So if the unemployment is above the natural rate 00:13:51.032 --> 00:13:52.990 align:middle line:84% of unemployment, that means inflation will tend 00:13:52.990 --> 00:13:54.960 align:middle line:90% to be below expected inflation. 00:13:54.960 --> 00:13:57.010 align:middle line:84% If expected inflation happens to be 00:13:57.010 --> 00:14:00.300 align:middle line:84% equal to log inflation, that means if unemployment 00:14:00.300 --> 00:14:02.050 align:middle line:84% is above the natural rate of unemployment, 00:14:02.050 --> 00:14:03.390 align:middle line:90% then inflation will be falling. 00:14:03.390 --> 00:14:07.017 align:middle line:90% 00:14:07.017 --> 00:14:07.600 align:middle line:90% Any questions? 00:14:07.600 --> 00:14:08.100 align:middle line:90% Good. 00:14:08.100 --> 00:14:13.800 align:middle line:84% You need to know this, how to derive these things. 00:14:13.800 --> 00:14:14.690 align:middle line:90% I mean, not so much-- 00:14:14.690 --> 00:14:16.190 align:middle line:84% Yeah, you should know how to derive. 00:14:16.190 --> 00:14:18.680 align:middle line:84% But you need to understand this relationship 00:14:18.680 --> 00:14:24.410 align:middle line:84% between the unemployment gap and inflation 00:14:24.410 --> 00:14:25.910 align:middle line:90% relative to expected inflation. 00:14:25.910 --> 00:14:27.390 align:middle line:90% Yep. 00:14:27.390 --> 00:14:32.277 align:middle line:84% AUDIENCE: Could you talk about uncut versus de-uncut inflation. 00:14:32.277 --> 00:14:33.860 align:middle line:84% RICARDO CABALLERO: Expected inflation. 00:14:33.860 --> 00:14:38.470 align:middle line:84% It's just a statement about a what 00:14:38.470 --> 00:14:43.320 align:middle line:84% is the model we have for expected inflation. 00:14:43.320 --> 00:14:49.250 align:middle line:84% So suppose we have the following model for expected inflation-- 00:14:49.250 --> 00:14:51.800 align:middle line:84% 1 minus theta-- theta is some number 00:14:51.800 --> 00:14:57.140 align:middle line:84% between 0 and 1-- times a constant inflation 00:14:57.140 --> 00:15:00.050 align:middle line:84% plus something that is a function of theta times 00:15:00.050 --> 00:15:03.380 align:middle line:90% whatever is previous inflation. 00:15:03.380 --> 00:15:06.140 align:middle line:84% Central banks try to set a target for the inflation rate. 00:15:06.140 --> 00:15:08.150 align:middle line:90% In the US it's around 2%. 00:15:08.150 --> 00:15:10.790 align:middle line:84% And ideally, people will tend to believe-- 00:15:10.790 --> 00:15:16.410 align:middle line:84% they may see an inflation that is above, say, 2%. 00:15:16.410 --> 00:15:20.510 align:middle line:84% But as long as people expect that to be undone 00:15:20.510 --> 00:15:24.110 align:middle line:84% in the next period, then inflation 00:15:24.110 --> 00:15:25.940 align:middle line:84% will say they are very well anchored. 00:15:25.940 --> 00:15:27.860 align:middle line:84% So that's a case in which very well anchored 00:15:27.860 --> 00:15:29.700 align:middle line:90% means theta equal to 0 here. 00:15:29.700 --> 00:15:34.100 align:middle line:84% And you always stick in there, in the case of the US, at 2%. 00:15:34.100 --> 00:15:37.590 align:middle line:84% And that's the way the economy is behaving right now. 00:15:37.590 --> 00:15:39.120 align:middle line:90% Inflation today is 5%. 00:15:39.120 --> 00:15:41.480 align:middle line:84% But if you ask people, what do you expect inflation 00:15:41.480 --> 00:15:43.670 align:middle line:90% to be two years from now? 00:15:43.670 --> 00:15:47.150 align:middle line:84% People tell you, look around 2%, 2.5% or so. 00:15:47.150 --> 00:15:50.930 align:middle line:84% An anchor expectation is when you don't have that anchor, 00:15:50.930 --> 00:15:53.488 align:middle line:84% that 2% that the Fed told you is whatever 00:15:53.488 --> 00:15:56.030 align:middle line:84% it was the previous inflation, that's what people extrapolate 00:15:56.030 --> 00:15:57.760 align:middle line:84% will be inflation for next period. 00:15:57.760 --> 00:16:00.563 align:middle line:84% And that's a lot harder when you get into an inflationary episode 00:16:00.563 --> 00:16:01.230 align:middle line:90% in that context. 00:16:01.230 --> 00:16:03.370 align:middle line:84% It's very difficult because you have 5%. 00:16:03.370 --> 00:16:05.560 align:middle line:84% People are still expecting 5% for next year. 00:16:05.560 --> 00:16:06.810 align:middle line:90% So you need to-- 00:16:06.810 --> 00:16:08.960 align:middle line:84% it's much harder to bring inflation down. 00:16:08.960 --> 00:16:11.460 align:middle line:84% You need to create much more unemployment to bring inflation 00:16:11.460 --> 00:16:14.070 align:middle line:90% back to the 2% target. 00:16:14.070 --> 00:16:15.780 align:middle line:90% That's what it means to anchor. 00:16:15.780 --> 00:16:18.630 align:middle line:84% So anchor means theta very close to 0. 00:16:18.630 --> 00:16:20.760 align:middle line:84% And anchor theta very close to 1. 00:16:20.760 --> 00:16:23.260 align:middle line:90% That's a formal definition. 00:16:23.260 --> 00:16:31.450 align:middle line:90% 00:16:31.450 --> 00:16:37.395 align:middle line:84% We then move to what I think is probably 00:16:37.395 --> 00:16:39.770 align:middle line:84% the most important model you'll see in this course, which 00:16:39.770 --> 00:16:44.410 align:middle line:84% is the IS-LM-PC, which is just the IS-LM plus the Phillips 00:16:44.410 --> 00:16:44.910 align:middle line:90% curve. 00:16:44.910 --> 00:16:47.360 align:middle line:84% And that allows us to talk about the short run, which 00:16:47.360 --> 00:16:50.270 align:middle line:84% is what we did in the IS-LM and then all the way 00:16:50.270 --> 00:16:52.520 align:middle line:90% to the medium run. 00:16:52.520 --> 00:16:54.020 align:middle line:84% A medium run understood, that's when 00:16:54.020 --> 00:16:57.200 align:middle line:84% you go back to the natural rate of unemployment, natural level 00:16:57.200 --> 00:17:00.730 align:middle line:90% of output, and so on. 00:17:00.730 --> 00:17:02.510 align:middle line:90% We got a banking crisis there. 00:17:02.510 --> 00:17:11.619 align:middle line:84% But that's-- This you may find useful. 00:17:11.619 --> 00:17:14.040 align:middle line:84% Here I was trying to explain the banking crisis. 00:17:14.040 --> 00:17:17.920 align:middle line:84% And I said, we have a model for that already. 00:17:17.920 --> 00:17:20.670 align:middle line:84% Remember, we had this x spreads in the investment function 00:17:20.670 --> 00:17:24.060 align:middle line:84% and said, well, you can think of a negative financial shock, 00:17:24.060 --> 00:17:26.369 align:middle line:84% something like a credit spread shock, 00:17:26.369 --> 00:17:29.820 align:middle line:84% as an increase in x, and that will shift IS to the left. 00:17:29.820 --> 00:17:33.390 align:middle line:90% OK, just saying. 00:17:33.390 --> 00:17:33.890 align:middle line:90% Good. 00:17:33.890 --> 00:17:37.590 align:middle line:90% 00:17:37.590 --> 00:17:41.430 align:middle line:84% IS-LM-PC model was just going back to the IS-LM model. 00:17:41.430 --> 00:17:44.730 align:middle line:84% We're going to simplify things by just assuming 00:17:44.730 --> 00:17:46.650 align:middle line:84% that the central bank sets the real interest 00:17:46.650 --> 00:17:50.870 align:middle line:84% rate and the real interest rate is that. 00:17:50.870 --> 00:17:54.810 align:middle line:84% And to that, we added a Phillips curve. 00:17:54.810 --> 00:17:56.790 align:middle line:84% But we didn't like that Phillips curve 00:17:56.790 --> 00:17:59.430 align:middle line:84% because we have everything is a function of output 00:17:59.430 --> 00:18:00.850 align:middle line:90% here and interest rate. 00:18:00.850 --> 00:18:04.780 align:middle line:84% And now we have inflation and the unemployment rate 00:18:04.780 --> 00:18:07.390 align:middle line:84% and so on-- yet another variable to carry around. 00:18:07.390 --> 00:18:13.320 align:middle line:84% So we went from the output gap to unemployment gap 00:18:13.320 --> 00:18:14.550 align:middle line:90% to an output gap. 00:18:14.550 --> 00:18:17.460 align:middle line:84% And we did that just by noticing that the output is 00:18:17.460 --> 00:18:23.610 align:middle line:84% equal to the labor force times 1 minus the unemployment rate. 00:18:23.610 --> 00:18:26.070 align:middle line:84% Similarly, you can define potential output 00:18:26.070 --> 00:18:29.340 align:middle line:84% or the natural output level as employment 00:18:29.340 --> 00:18:31.690 align:middle line:84% times 1 minus the natural rate of unemployment. 00:18:31.690 --> 00:18:36.720 align:middle line:84% Subtract these 2 and you get that the output 00:18:36.720 --> 00:18:41.510 align:middle line:84% gap is equal to minus L times the unemployment gap. 00:18:41.510 --> 00:18:44.698 align:middle line:84% And so we replace this for that expression divided by L 00:18:44.698 --> 00:18:46.240 align:middle line:84% and we end up with the Phillips curve 00:18:46.240 --> 00:18:48.910 align:middle line:84% written in the form of an increasing 00:18:48.910 --> 00:18:50.360 align:middle line:90% function of the output gap. 00:18:50.360 --> 00:18:52.540 align:middle line:84% So when the output gap is positive, 00:18:52.540 --> 00:18:55.370 align:middle line:84% then inflation will exceed expected inflation. 00:18:55.370 --> 00:18:58.120 align:middle line:84% If expected inflation is an anchor, that is, 00:18:58.120 --> 00:19:00.850 align:middle line:84% expected inflation is equal to log inflation, then 00:19:00.850 --> 00:19:02.410 align:middle line:84% that means that a positive output 00:19:02.410 --> 00:19:05.870 align:middle line:84% gap leads to an increase in inflation, in inflation rate. 00:19:05.870 --> 00:19:09.130 align:middle line:90% 00:19:09.130 --> 00:19:11.500 align:middle line:90% We look at an example here. 00:19:11.500 --> 00:19:12.573 align:middle line:90% This is the type of-- 00:19:12.573 --> 00:19:14.740 align:middle line:84% now we're going to have the real interest rate here. 00:19:14.740 --> 00:19:17.452 align:middle line:84% Just makes it simpler to think about central monetary policy 00:19:17.452 --> 00:19:18.910 align:middle line:84% in terms of the real interest rate. 00:19:18.910 --> 00:19:21.130 align:middle line:84% Otherwise, too many things move at once. 00:19:21.130 --> 00:19:23.450 align:middle line:84% So this is what we have done for quiz 1. 00:19:23.450 --> 00:19:26.260 align:middle line:84% Here you have some particular equilibrium, IS-LM. 00:19:26.260 --> 00:19:31.300 align:middle line:84% With this real interest rate, we got some equilibrium output 00:19:31.300 --> 00:19:33.790 align:middle line:90% equal to y. 00:19:33.790 --> 00:19:36.310 align:middle line:84% The contribution of this block of the course 00:19:36.310 --> 00:19:38.710 align:middle line:84% is that now we need to also check 00:19:38.710 --> 00:19:42.430 align:middle line:84% whether this is y is consistent with potential output 00:19:42.430 --> 00:19:44.640 align:middle line:84% or not, or with natural level of output. 00:19:44.640 --> 00:19:50.290 align:middle line:84% And that for that we need to see whether this level of output is, 00:19:50.290 --> 00:19:53.480 align:middle line:84% again, is above or below the natural rate of output. 00:19:53.480 --> 00:19:56.120 align:middle line:84% And for that, we need to look at the Phillips curve. 00:19:56.120 --> 00:19:59.020 align:middle line:84% And in this particular case, that's 00:19:59.020 --> 00:20:02.170 align:middle line:84% not the case because output is above the natural rate 00:20:02.170 --> 00:20:03.040 align:middle line:90% of output. 00:20:03.040 --> 00:20:05.710 align:middle line:84% You put now, given that observation, 00:20:05.710 --> 00:20:09.280 align:middle line:84% you draw here the Phillips curve. 00:20:09.280 --> 00:20:11.860 align:middle line:84% You know that because output is above the natural rate 00:20:11.860 --> 00:20:14.950 align:middle line:84% of output, the natural level of output, that means inflation 00:20:14.950 --> 00:20:16.760 align:middle line:90% is above expected inflation. 00:20:16.760 --> 00:20:18.640 align:middle line:84% If expected inflation happens to be 00:20:18.640 --> 00:20:20.500 align:middle line:84% an anchor equal to pi minus one, that 00:20:20.500 --> 00:20:24.010 align:middle line:84% means that at this output gap, that there 00:20:24.010 --> 00:20:26.050 align:middle line:90% is an inflation that is rising. 00:20:26.050 --> 00:20:31.490 align:middle line:84% Now, inflation rising means the central bank will have to react. 00:20:31.490 --> 00:20:33.940 align:middle line:84% And so you have to do something up here. 00:20:33.940 --> 00:20:36.430 align:middle line:90% You need to bring output down. 00:20:36.430 --> 00:20:40.610 align:middle line:84% And how can you bring output down? 00:20:40.610 --> 00:20:44.130 align:middle line:84% So this economy is engaging in an inflationary spiral, 00:20:44.130 --> 00:20:46.500 align:middle line:84% actually, given this model of expectation. 00:20:46.500 --> 00:20:47.615 align:middle line:90% How do you stop that? 00:20:47.615 --> 00:20:55.378 align:middle line:90% 00:20:55.378 --> 00:20:56.670 align:middle line:90% AUDIENCE: Raise interest rates. 00:20:56.670 --> 00:20:57.750 align:middle line:84% RICARDO CABALLERO: If you are the Fed, 00:20:57.750 --> 00:20:59.850 align:middle line:84% and you when-- you raise interest rates, no? 00:20:59.850 --> 00:21:01.600 align:middle line:84% Because you need to bring them back. 00:21:01.600 --> 00:21:03.530 align:middle line:84% So the equilibrium level of output, 00:21:03.530 --> 00:21:09.720 align:middle line:84% you increase the real rate up to a point in which the equilibrium 00:21:09.720 --> 00:21:13.830 align:middle line:84% level of output is equal to the natural rate of output. 00:21:13.830 --> 00:21:15.910 align:middle line:84% And you may have to do more than that. 00:21:15.910 --> 00:21:17.910 align:middle line:84% If inflation was an anchor and you find yourself 00:21:17.910 --> 00:21:21.030 align:middle line:84% with 5% inflation, you may have to temporarily actually, 00:21:21.030 --> 00:21:23.260 align:middle line:84% to bring inflation back down to 2%, 00:21:23.260 --> 00:21:25.800 align:middle line:84% you may have to overshoot, raise interest rates a lot, 00:21:25.800 --> 00:21:28.270 align:middle line:84% generate a negative output gap for a while. 00:21:28.270 --> 00:21:32.070 align:middle line:84% And then once you reach the level of inflation you like, 00:21:32.070 --> 00:21:36.440 align:middle line:84% the 2%, then you can go back to the natural level of output. 00:21:36.440 --> 00:21:39.630 align:middle line:84% OK, so that's the reason central banks worry 00:21:39.630 --> 00:21:41.310 align:middle line:84% a lot about unanchored expectations 00:21:41.310 --> 00:21:43.852 align:middle line:84% because then they know that they find themselves on inflation 00:21:43.852 --> 00:21:45.420 align:middle line:84% above their target, it's not going 00:21:45.420 --> 00:21:47.950 align:middle line:84% to be enough to bring the output gap to 0. 00:21:47.950 --> 00:21:50.190 align:middle line:84% They're going to have to overshoot in the way down 00:21:50.190 --> 00:21:52.650 align:middle line:84% in order to reanchor expectations-- well, 00:21:52.650 --> 00:21:57.030 align:middle line:84% in order to bring inflation back down to the target of 2%. 00:21:57.030 --> 00:22:00.330 align:middle line:84% But in any event, even if inflation expectation's well 00:22:00.330 --> 00:22:02.577 align:middle line:84% anchored, you still have to bring output down 00:22:02.577 --> 00:22:04.410 align:middle line:84% because at the very least, you need to close 00:22:04.410 --> 00:22:06.120 align:middle line:90% this positive output gap. 00:22:06.120 --> 00:22:09.570 align:middle line:84% And that, if you are the Fed in the US or any central bank, 00:22:09.570 --> 00:22:11.620 align:middle line:84% you do it by increasing the real interest rate. 00:22:11.620 --> 00:22:13.558 align:middle line:84% Now, in practice, central banks really 00:22:13.558 --> 00:22:15.100 align:middle line:84% don't control the real interest rate, 00:22:15.100 --> 00:22:16.517 align:middle line:84% control the nominal interest rate. 00:22:16.517 --> 00:22:18.750 align:middle line:84% So there is a little fight there between inflation 00:22:18.750 --> 00:22:20.980 align:middle line:84% and what they do to the nominal interest rate. 00:22:20.980 --> 00:22:23.440 align:middle line:84% But let's ignore that complication for now. 00:22:23.440 --> 00:22:32.940 align:middle line:84% Now, suppose that the Fed is in vacation and so somebody someone 00:22:32.940 --> 00:22:37.110 align:middle line:84% else decides in the government decides that, no, 00:22:37.110 --> 00:22:39.670 align:middle line:84% we cannot have this very high level of inflation. 00:22:39.670 --> 00:22:40.760 align:middle line:90% So what else could you do? 00:22:40.760 --> 00:22:43.730 align:middle line:90% 00:22:43.730 --> 00:22:44.850 align:middle line:90% And you're not the Fed. 00:22:44.850 --> 00:22:47.940 align:middle line:90% The Fed is in a vacation. 00:22:47.940 --> 00:22:51.540 align:middle line:90% Who else can make policy? 00:22:51.540 --> 00:22:54.040 align:middle line:84% The government, the central government, the Treasury, and so 00:22:54.040 --> 00:22:54.580 align:middle line:90% on. 00:22:54.580 --> 00:22:56.950 align:middle line:84% What is the instrument they have? 00:22:56.950 --> 00:22:59.097 align:middle line:90% What do they need to do? 00:22:59.097 --> 00:23:00.930 align:middle line:84% The problem they have is output is too high. 00:23:00.930 --> 00:23:03.210 align:middle line:84% And that's what is leading to lots of inflation. 00:23:03.210 --> 00:23:06.435 align:middle line:84% So what do you think they should do? 00:23:06.435 --> 00:23:09.503 align:middle line:90% 00:23:09.503 --> 00:23:10.920 align:middle line:84% AUDIENCE: Cut government spending. 00:23:10.920 --> 00:23:12.000 align:middle line:84% RICARDO CABALLERO: Cut government spending, 00:23:12.000 --> 00:23:13.910 align:middle line:84% raise taxes, something of that kind. 00:23:13.910 --> 00:23:15.900 align:middle line:84% But you need a fiscal contraction 00:23:15.900 --> 00:23:18.460 align:middle line:84% because that will bring the IS down. 00:23:18.460 --> 00:23:22.260 align:middle line:84% And so the equilibrium output will be lower. 00:23:22.260 --> 00:23:24.330 align:middle line:84% So that's an alternative you have. 00:23:24.330 --> 00:23:25.205 align:middle line:90% You should know this. 00:23:25.205 --> 00:23:32.480 align:middle line:90% 00:23:32.480 --> 00:23:36.920 align:middle line:84% And here I just did what we just discussed, just in steps. 00:23:36.920 --> 00:23:38.600 align:middle line:90% These things happen slowly. 00:23:38.600 --> 00:23:41.400 align:middle line:84% The Fed doesn't hike interest rates in one shot and so on. 00:23:41.400 --> 00:23:44.910 align:middle line:84% It takes a while before you get to the final equilibrium. 00:23:44.910 --> 00:23:54.384 align:middle line:90% 00:23:54.384 --> 00:23:56.740 align:middle line:84% I showed you the deflationary spiral. 00:23:56.740 --> 00:24:02.000 align:middle line:84% I said, sometimes things can get very complicated because you 00:24:02.000 --> 00:24:03.390 align:middle line:90% may hit the 0 lower bound. 00:24:03.390 --> 00:24:06.410 align:middle line:84% The Fed can bring the nominal interest rate to 0. 00:24:06.410 --> 00:24:08.715 align:middle line:84% But if inflation is already low, that 00:24:08.715 --> 00:24:10.340 align:middle line:84% may not give you the real interest rate 00:24:10.340 --> 00:24:12.530 align:middle line:84% you need in order to get output equal to the natural rate 00:24:12.530 --> 00:24:13.160 align:middle line:90% of output. 00:24:13.160 --> 00:24:15.620 align:middle line:84% Here was one example in which you 00:24:15.620 --> 00:24:18.050 align:middle line:84% need a negative real interest rate to get output 00:24:18.050 --> 00:24:20.150 align:middle line:84% to be equal to the natural rate of output. 00:24:20.150 --> 00:24:25.730 align:middle line:84% But that may not happen because you hit the 0 lower bound. 00:24:25.730 --> 00:24:28.520 align:middle line:84% And so at that point, the problem you have is that-- 00:24:28.520 --> 00:24:31.940 align:middle line:84% and that was a tragedy of Japan for so long-- 00:24:31.940 --> 00:24:38.060 align:middle line:84% is that not only you cannot bring the interest rate, 00:24:38.060 --> 00:24:39.560 align:middle line:84% the nominal interest rate below 0, 00:24:39.560 --> 00:24:41.780 align:middle line:84% but you start getting into deflationary. 00:24:41.780 --> 00:24:43.970 align:middle line:84% Inflation below expectation, and the expectation 00:24:43.970 --> 00:24:45.920 align:middle line:84% goes to number very close to 0 because 00:24:45.920 --> 00:24:48.220 align:middle line:84% of an anchor deflationary expectations. 00:24:48.220 --> 00:24:51.155 align:middle line:84% Then you start getting negative expected inflation. 00:24:51.155 --> 00:24:53.030 align:middle line:84% And when you get negative expected inflation, 00:24:53.030 --> 00:24:55.620 align:middle line:84% even if you are at the 0 lower bound in the nominal interest 00:24:55.620 --> 00:24:57.610 align:middle line:84% rate, that means a positive real interest rate. 00:24:57.610 --> 00:24:59.400 align:middle line:84% So effectively, you are increasing interest rate 00:24:59.400 --> 00:25:00.130 align:middle line:90% at the same time. 00:25:00.130 --> 00:25:02.820 align:middle line:84% And that can be a very complicated thing to get out of. 00:25:02.820 --> 00:25:06.180 align:middle line:84% Again, that's what happened to Japan for a long time. 00:25:06.180 --> 00:25:08.610 align:middle line:84% What would you do as a government 00:25:08.610 --> 00:25:10.475 align:middle line:84% if you fall into a situation like that? 00:25:10.475 --> 00:25:13.870 align:middle line:90% 00:25:13.870 --> 00:25:16.917 align:middle line:90% And Japan did a lot of that. 00:25:16.917 --> 00:25:18.250 align:middle line:90% Well, you can do lots of things. 00:25:18.250 --> 00:25:21.150 align:middle line:84% But in particular, the kind of things 00:25:21.150 --> 00:25:23.070 align:middle line:90% you know, what would you do? 00:25:23.070 --> 00:25:26.430 align:middle line:84% If you are in a situation like this in which the 0 lower bound 00:25:26.430 --> 00:25:29.280 align:middle line:84% is binding and inflation is actually falling? 00:25:29.280 --> 00:25:32.460 align:middle line:84% Here I had a benign case in which inflation expectation 00:25:32.460 --> 00:25:33.700 align:middle line:90% was well anchored. 00:25:33.700 --> 00:25:35.140 align:middle line:84% That's not what happened to Japan. 00:25:35.140 --> 00:25:37.810 align:middle line:84% After they experience a long period of deflationary forces, 00:25:37.810 --> 00:25:41.470 align:middle line:84% then people began to expect more deflation, more deflation, 00:25:41.470 --> 00:25:42.160 align:middle line:90% and so on. 00:25:42.160 --> 00:25:44.405 align:middle line:90% So what else can you do? 00:25:44.405 --> 00:25:52.630 align:middle line:90% 00:25:52.630 --> 00:25:54.600 align:middle line:90% Let me give you a hint. 00:25:54.600 --> 00:25:56.730 align:middle line:84% Japan is one of the countries at the highest 00:25:56.730 --> 00:25:59.330 align:middle line:90% levels of public debt. 00:25:59.330 --> 00:26:03.420 align:middle line:84% How do you accumulate public debt? 00:26:03.420 --> 00:26:05.590 align:middle line:90% Yeah, you need to borrow a lot. 00:26:05.590 --> 00:26:07.080 align:middle line:90% You have big fiscal deficit. 00:26:07.080 --> 00:26:08.780 align:middle line:84% So that's the way you can fight this. 00:26:08.780 --> 00:26:11.070 align:middle line:84% You can shift IS to the right by having 00:26:11.070 --> 00:26:12.680 align:middle line:90% an expansionary fiscal policy. 00:26:12.680 --> 00:26:14.130 align:middle line:84% That's the only tool really have. 00:26:14.130 --> 00:26:16.230 align:middle line:84% You lose the power of monetary policy 00:26:16.230 --> 00:26:17.550 align:middle line:90% against the 0 lower bound. 00:26:17.550 --> 00:26:19.150 align:middle line:84% But you still have fiscal policy. 00:26:19.150 --> 00:26:20.650 align:middle line:84% And they did a lot of fiscal policy. 00:26:20.650 --> 00:26:27.880 align:middle line:90% 00:26:27.880 --> 00:26:30.740 align:middle line:90% No, not interesting. 00:26:30.740 --> 00:26:34.510 align:middle line:90% This is interesting. 00:26:34.510 --> 00:26:36.670 align:middle line:84% This is a different kind of shock. 00:26:36.670 --> 00:26:38.920 align:middle line:84% Suppose you are at your medium-run equilibrium 00:26:38.920 --> 00:26:41.110 align:middle line:84% and then all of a sudden mark ups 00:26:41.110 --> 00:26:44.440 align:middle line:84% go up, perhaps, for example, because the price of oil 00:26:44.440 --> 00:26:46.210 align:middle line:84% went up a lot and something like that. 00:26:46.210 --> 00:26:50.920 align:middle line:84% So then that's a different kind of shock from the previous one, 00:26:50.920 --> 00:26:53.065 align:middle line:84% from any fiscal shock or anything like that. 00:26:53.065 --> 00:26:54.190 align:middle line:90% That's an aggregate demand. 00:26:54.190 --> 00:26:55.648 align:middle line:84% This is an aggregate supply problem 00:26:55.648 --> 00:26:58.630 align:middle line:84% because the first thing I know of a permanent at least 00:26:58.630 --> 00:27:01.810 align:middle line:84% change in m is that the natural rate of unemployment 00:27:01.810 --> 00:27:03.840 align:middle line:90% has to rise. 00:27:03.840 --> 00:27:06.000 align:middle line:84% If the natural rate of unemployment has to rise, 00:27:06.000 --> 00:27:10.580 align:middle line:84% that means my Phillips curve will shift now. 00:27:10.580 --> 00:27:12.942 align:middle line:84% In that particular case, I know the Phillips curve 00:27:12.942 --> 00:27:13.900 align:middle line:90% will shift to the left. 00:27:13.900 --> 00:27:15.120 align:middle line:90% How do I know that? 00:27:15.120 --> 00:27:18.590 align:middle line:84% Well, because I know that the natural rate of unemployment 00:27:18.590 --> 00:27:21.800 align:middle line:84% went up, which means that the natural rate, natural level 00:27:21.800 --> 00:27:23.600 align:middle line:90% of output has to come down. 00:27:23.600 --> 00:27:25.610 align:middle line:84% And the natural level of output coming down 00:27:25.610 --> 00:27:29.060 align:middle line:84% means simply that the level at which expected inflation 00:27:29.060 --> 00:27:33.050 align:middle line:84% and inflation are equal happens at a lower level of output. 00:27:33.050 --> 00:27:34.780 align:middle line:84% So the Phillips curve moves to the left. 00:27:34.780 --> 00:27:36.860 align:middle line:84% So suppose you were in this equilibrium. 00:27:36.860 --> 00:27:39.300 align:middle line:84% Here I'm doing for the case of an anchor expectations. 00:27:39.300 --> 00:27:44.030 align:middle line:84% But the same logic goes for the case of anchor expectations. 00:27:44.030 --> 00:27:46.910 align:middle line:84% So suppose you were at some equilibrium like this. 00:27:46.910 --> 00:27:48.390 align:middle line:84% What's your medium run equilibrium? 00:27:48.390 --> 00:27:50.630 align:middle line:84% But now the price of energy goes up a lot 00:27:50.630 --> 00:27:53.290 align:middle line:84% and you expect that to last for a while. 00:27:53.290 --> 00:27:55.530 align:middle line:84% That means the Phillips curve moves up. 00:27:55.530 --> 00:27:58.490 align:middle line:84% So that means that if output, with output at this level, 00:27:58.490 --> 00:28:01.970 align:middle line:84% now you have a problem because you start getting inflation out 00:28:01.970 --> 00:28:04.970 align:middle line:84% of this, because this level of output 00:28:04.970 --> 00:28:08.390 align:middle line:84% is too high relative to the new level of the natural-- 00:28:08.390 --> 00:28:10.960 align:middle line:90% the new natural level of output. 00:28:10.960 --> 00:28:12.720 align:middle line:84% So you have a positive output gap. 00:28:12.720 --> 00:28:14.520 align:middle line:84% Positive output gap means inflation 00:28:14.520 --> 00:28:16.080 align:middle line:90% above expected inflation. 00:28:16.080 --> 00:28:17.640 align:middle line:84% If you have an anchor expectations, 00:28:17.640 --> 00:28:22.410 align:middle line:84% means inflation starts rising up. 00:28:22.410 --> 00:28:25.470 align:middle line:84% So that means the Fed now needs to react to that 00:28:25.470 --> 00:28:28.110 align:middle line:84% and needs to tighten interest rate in order 00:28:28.110 --> 00:28:32.370 align:middle line:84% to go to the new level of, natural level of output. 00:28:32.370 --> 00:28:34.320 align:middle line:90% And that's the response to that. 00:28:34.320 --> 00:28:36.060 align:middle line:90% But if the Fed does not react-- 00:28:36.060 --> 00:28:37.890 align:middle line:84% and a little bit of this is what happened. 00:28:37.890 --> 00:28:40.252 align:middle line:84% We had some supply shocks and so on that 00:28:40.252 --> 00:28:42.960 align:middle line:84% were considered to be temporary, well, they weren't as temporary. 00:28:42.960 --> 00:28:44.680 align:middle line:90% So there was no reaction. 00:28:44.680 --> 00:28:47.310 align:middle line:84% But it turns out that they lasted a lot longer 00:28:47.310 --> 00:28:48.910 align:middle line:90% than the Fed expected. 00:28:48.910 --> 00:28:53.760 align:middle line:90% And so now they had to catch up. 00:28:53.760 --> 00:28:57.450 align:middle line:84% That was part of the reason we got into a high inflation 00:28:57.450 --> 00:28:58.920 align:middle line:90% episode. 00:28:58.920 --> 00:29:02.360 align:middle line:84% That was the main reason in Europe. 00:29:02.360 --> 00:29:05.930 align:middle line:84% The US is a mixture of aggregate demand, lots of fiscal policy 00:29:05.930 --> 00:29:07.760 align:middle line:90% and so on. 00:29:07.760 --> 00:29:12.150 align:middle line:84% And supply side in Europe was very much a story of this kind. 00:29:12.150 --> 00:29:20.120 align:middle line:90% 00:29:20.120 --> 00:29:22.880 align:middle line:84% Well, a financial panic, you need 00:29:22.880 --> 00:29:26.090 align:middle line:84% to offset it with a decline in real interest rates. 00:29:26.090 --> 00:29:28.340 align:middle line:84% And a little bit of that has been happening. 00:29:28.340 --> 00:29:30.530 align:middle line:84% It's not the Fed that has cut the rates, 00:29:30.530 --> 00:29:32.990 align:middle line:84% but the markets have anticipated the Fed will not 00:29:32.990 --> 00:29:37.210 align:middle line:84% raise interest rates as much as they expected before we 00:29:37.210 --> 00:29:39.340 align:middle line:90% got into this banking mess. 00:29:39.340 --> 00:29:42.730 align:middle line:84% So we had already studied the short run, the medium run. 00:29:42.730 --> 00:29:45.100 align:middle line:84% And now we want to look at the long run. 00:29:45.100 --> 00:29:47.930 align:middle line:84% And that's what economic growth is about-- 00:29:47.930 --> 00:29:50.570 align:middle line:84% economic growth theory and facts and so on. 00:29:50.570 --> 00:29:54.770 align:middle line:90% 00:29:54.770 --> 00:29:56.450 align:middle line:90% Let me go to-- 00:29:56.450 --> 00:29:59.450 align:middle line:84% So one of the things I highlighted 00:29:59.450 --> 00:30:03.440 align:middle line:84% is that we tend to see, among countries 00:30:03.440 --> 00:30:10.060 align:middle line:84% that are fairly similar along the education and variables 00:30:10.060 --> 00:30:11.050 align:middle line:90% like that-- 00:30:11.050 --> 00:30:14.350 align:middle line:84% systems, economic systems and political systems and so 00:30:14.350 --> 00:30:16.240 align:middle line:84% on-- you tend to find relationships 00:30:16.240 --> 00:30:20.722 align:middle line:84% like this, that is countries with a lower per capita income 00:30:20.722 --> 00:30:22.180 align:middle line:84% at the beginning of the sample tend 00:30:22.180 --> 00:30:23.630 align:middle line:90% to grow faster in the sample. 00:30:23.630 --> 00:30:25.150 align:middle line:84% And that captures very much the idea 00:30:25.150 --> 00:30:27.430 align:middle line:84% that there is a convergence, there 00:30:27.430 --> 00:30:31.210 align:middle line:84% is a force towards convergence of income per capita, 00:30:31.210 --> 00:30:33.802 align:middle line:90% if you will. 00:30:33.802 --> 00:30:35.760 align:middle line:84% That's another illustration of that phenomenon. 00:30:35.760 --> 00:30:38.340 align:middle line:90% Lots of dispersion here. 00:30:38.340 --> 00:30:42.470 align:middle line:84% 70 years later, a lot less dispersion. 00:30:42.470 --> 00:30:45.770 align:middle line:84% But we also said that some countries do not match that. 00:30:45.770 --> 00:30:52.400 align:middle line:84% And but we focus most of what we did in growth on understanding 00:30:52.400 --> 00:30:54.830 align:middle line:84% this process, the process of convergence 00:30:54.830 --> 00:30:57.420 align:middle line:84% and how it happened without technological progress 00:30:57.420 --> 00:30:58.200 align:middle line:90% and so on. 00:30:58.200 --> 00:31:00.890 align:middle line:84% And then we spent a little bit of a lecture, 00:31:00.890 --> 00:31:05.510 align:middle line:84% say at most 10 points worth in a quiz, or seven points, 00:31:05.510 --> 00:31:08.447 align:middle line:84% talking about anomalies and things like that-- 00:31:08.447 --> 00:31:09.780 align:middle line:90% 5 points or something like that. 00:31:09.780 --> 00:31:13.060 align:middle line:90% 00:31:13.060 --> 00:31:17.210 align:middle line:84% So the key object here, one of the key objects-- 00:31:17.210 --> 00:31:19.220 align:middle line:90% there are a couple-- 00:31:19.220 --> 00:31:21.215 align:middle line:84% but one of them was, well, now we 00:31:21.215 --> 00:31:23.590 align:middle line:84% need to be a little bit more serious about the production 00:31:23.590 --> 00:31:24.340 align:middle line:90% function, we said. 00:31:24.340 --> 00:31:27.550 align:middle line:84% Because for the short run, it's OK to take capital 00:31:27.550 --> 00:31:29.073 align:middle line:84% as given and just worry about most 00:31:29.073 --> 00:31:30.490 align:middle line:84% of the fluctuations in output will 00:31:30.490 --> 00:31:32.900 align:middle line:84% come from fluctuations in employment. 00:31:32.900 --> 00:31:35.620 align:middle line:84% That's not so over long periods of time. 00:31:35.620 --> 00:31:37.960 align:middle line:84% Capital accumulation plays a huge role. 00:31:37.960 --> 00:31:41.080 align:middle line:84% And so we need to be explicit about the fact 00:31:41.080 --> 00:31:43.690 align:middle line:84% that capital matters a lot for production. 00:31:43.690 --> 00:31:48.350 align:middle line:84% And so we postulated a production function 00:31:48.350 --> 00:31:53.938 align:middle line:84% like this, output as increasing function of capital and labor. 00:31:53.938 --> 00:31:55.730 align:middle line:84% And now we said for this part of the course 00:31:55.730 --> 00:31:57.563 align:middle line:84% we're not going to worry about unemployment. 00:31:57.563 --> 00:31:59.720 align:middle line:84% And so on, employment, labor force population, 00:31:59.720 --> 00:32:01.880 align:middle line:84% they're are all the same for us here, 00:32:01.880 --> 00:32:03.390 align:middle line:90% for this part of the course. 00:32:03.390 --> 00:32:05.300 align:middle line:84% And then they said this production function 00:32:05.300 --> 00:32:08.590 align:middle line:90% has some important properties. 00:32:08.590 --> 00:32:11.690 align:middle line:84% One, it has constant returns to scale. 00:32:11.690 --> 00:32:12.980 align:middle line:90% Things change quite a bit. 00:32:12.980 --> 00:32:14.813 align:middle line:84% If you don't have constant returns to scale. 00:32:14.813 --> 00:32:17.260 align:middle line:84% So we have constant returns to scale, which means-- 00:32:17.260 --> 00:32:18.430 align:middle line:90% you should know this-- 00:32:18.430 --> 00:32:22.360 align:middle line:84% that if you scale all output, all inputs, all the factors 00:32:22.360 --> 00:32:24.640 align:middle line:84% of production by the same factor, 00:32:24.640 --> 00:32:27.980 align:middle line:84% output also rises by the same factor. 00:32:27.980 --> 00:32:33.860 align:middle line:84% OK, so our production function we use a lot was Cobb-Douglas, 00:32:33.860 --> 00:32:37.860 align:middle line:84% output equals square root of k, times square root of n. 00:32:37.860 --> 00:32:39.960 align:middle line:84% Well, the sum of those exponents is 1. 00:32:39.960 --> 00:32:44.780 align:middle line:84% So that's a production function with constant returns to scale. 00:32:44.780 --> 00:32:48.440 align:middle line:84% Anything that has the exponents add up to 1, then that's 00:32:48.440 --> 00:32:50.990 align:middle line:84% a constant returns to scale technology. 00:32:50.990 --> 00:32:56.540 align:middle line:84% But importantly, it also has decreasing returns with each 00:32:56.540 --> 00:32:58.350 align:middle line:90% of its factors of production. 00:32:58.350 --> 00:33:03.620 align:middle line:84% That means rather than moving both factors of production up, 00:33:03.620 --> 00:33:04.950 align:middle line:90% you move only one. 00:33:04.950 --> 00:33:07.400 align:middle line:84% Well, you're going to increase output, but just 00:33:07.400 --> 00:33:08.985 align:middle line:84% keep increasing that factor alone, 00:33:08.985 --> 00:33:11.360 align:middle line:84% you're going to increase output by less and less and less 00:33:11.360 --> 00:33:12.860 align:middle line:84% and less because essentially, it has 00:33:12.860 --> 00:33:16.130 align:middle line:84% fewer and fewer of the other factor of production 00:33:16.130 --> 00:33:17.040 align:middle line:90% to work with. 00:33:17.040 --> 00:33:20.520 align:middle line:84% And so that's decreasing returns to capital or labor. 00:33:20.520 --> 00:33:22.820 align:middle line:84% I mean, if you fix the other factor of production, 00:33:22.820 --> 00:33:23.510 align:middle line:90% you move up. 00:33:23.510 --> 00:33:26.240 align:middle line:84% It's going to increase at a decreasing rate. 00:33:26.240 --> 00:33:29.060 align:middle line:84% So one normalization that we started with 00:33:29.060 --> 00:33:32.420 align:middle line:84% was, well, a scaling factor could be 00:33:32.420 --> 00:33:35.510 align:middle line:90% population, 1 over population. 00:33:35.510 --> 00:33:36.840 align:middle line:90% That's a scaling factor. 00:33:36.840 --> 00:33:40.610 align:middle line:84% And if I do that, I multiply everything by 1 over n. 00:33:40.610 --> 00:33:46.070 align:middle line:84% We go into output per person is an increasing function of-- 00:33:46.070 --> 00:33:48.710 align:middle line:84% an increasing function, but at a decreasing rate-- 00:33:48.710 --> 00:33:52.090 align:middle line:90% of capital per person. 00:33:52.090 --> 00:33:54.580 align:middle line:90% We plot that function here. 00:33:54.580 --> 00:33:57.680 align:middle line:84% And this is increasing, but it's concave. 00:33:57.680 --> 00:34:02.740 align:middle line:84% That shows the decreasing returns of capital. 00:34:02.740 --> 00:34:04.335 align:middle line:90% And we got that function there. 00:34:04.335 --> 00:34:10.940 align:middle line:90% 00:34:10.940 --> 00:34:15.909 align:middle line:84% So when you move in this, you can increase output per person, 00:34:15.909 --> 00:34:19.580 align:middle line:84% per person by simply increasing capital per person. 00:34:19.580 --> 00:34:22.900 align:middle line:84% And the more you increase capital per person, 00:34:22.900 --> 00:34:25.969 align:middle line:84% output will increase more and more but at a decreasing rate. 00:34:25.969 --> 00:34:29.050 align:middle line:84% You can see that moving the distance between A and B 00:34:29.050 --> 00:34:31.449 align:middle line:84% is the same as the distance between C and D. However, 00:34:31.449 --> 00:34:34.330 align:middle line:84% the increase in output, when you go from A to B, 00:34:34.330 --> 00:34:37.870 align:middle line:84% is enormous when compared with the increase in output 00:34:37.870 --> 00:34:39.580 align:middle line:84% that you get from the increase in capital 00:34:39.580 --> 00:34:44.860 align:middle line:84% over per person from C to D, decreasing returns. 00:34:44.860 --> 00:34:49.030 align:middle line:84% There is another way of increasing output per person, 00:34:49.030 --> 00:34:51.980 align:middle line:84% which is with technological progress, 00:34:51.980 --> 00:34:54.820 align:middle line:84% with the function, F, is shifting up over time. 00:34:54.820 --> 00:34:57.880 align:middle line:84% And we split the two main lectures in growth 00:34:57.880 --> 00:35:02.450 align:middle line:84% into part one, which shut down the second channel. 00:35:02.450 --> 00:35:05.080 align:middle line:84% And then the second important lecture 00:35:05.080 --> 00:35:09.140 align:middle line:90% here, we focus on this channel. 00:35:09.140 --> 00:35:10.650 align:middle line:90% So that's what we. 00:35:10.650 --> 00:35:15.070 align:middle line:90% 00:35:15.070 --> 00:35:18.340 align:middle line:84% So let's go to when I shut down this channel for now 00:35:18.340 --> 00:35:22.240 align:middle line:84% and focus on the case without technological progress first. 00:35:22.240 --> 00:35:24.410 align:middle line:90% So we put things together. 00:35:24.410 --> 00:35:26.660 align:middle line:84% We said this comes from the previous lecture. 00:35:26.660 --> 00:35:33.160 align:middle line:84% We can write output per person as an increasing function 00:35:33.160 --> 00:35:36.930 align:middle line:90% of capital per person. 00:35:36.930 --> 00:35:39.400 align:middle line:90% Second key equation is-- 00:35:39.400 --> 00:35:41.650 align:middle line:84% well, this is a property and it has to be, 00:35:41.650 --> 00:35:43.840 align:middle line:84% if you are in a closed economy, no government 00:35:43.840 --> 00:35:45.760 align:middle line:90% expenditure or anything. 00:35:45.760 --> 00:35:46.720 align:middle line:90% We could add that. 00:35:46.720 --> 00:35:49.050 align:middle line:84% But it's not important for the message. 00:35:49.050 --> 00:35:50.600 align:middle line:90% Then investment has to be equal. 00:35:50.600 --> 00:35:52.600 align:middle line:84% So investment is going to be very important here 00:35:52.600 --> 00:35:56.060 align:middle line:84% because what will make the capital stock grow. 00:35:56.060 --> 00:35:58.310 align:middle line:84% But there has to be funding for that. 00:35:58.310 --> 00:35:59.960 align:middle line:84% And the funding come from savings. 00:35:59.960 --> 00:36:03.070 align:middle line:84% And we simplify things by assuming that the saving 00:36:03.070 --> 00:36:05.470 align:middle line:84% function is just proportional to the level of output, 00:36:05.470 --> 00:36:07.750 align:middle line:84% which is reasonable when you think about long run. 00:36:07.750 --> 00:36:09.400 align:middle line:90% All these things scale up. 00:36:09.400 --> 00:36:11.270 align:middle line:84% When you're thinking about very short term, 00:36:11.270 --> 00:36:14.980 align:middle line:84% no, we have some constants and so on floating around. 00:36:14.980 --> 00:36:18.100 align:middle line:84% But over the long run, things do scale up. 00:36:18.100 --> 00:36:21.117 align:middle line:84% And so we can write investment in equilibrium. 00:36:21.117 --> 00:36:22.700 align:middle line:84% Investment has to be equal to savings. 00:36:22.700 --> 00:36:24.800 align:middle line:84% Saving is proportional to output. 00:36:24.800 --> 00:36:27.670 align:middle line:84% So we get that investment in this economy 00:36:27.670 --> 00:36:30.280 align:middle line:90% is increasing in output. 00:36:30.280 --> 00:36:33.020 align:middle line:84% This is a constant somewhere between 0 and 1. 00:36:33.020 --> 00:36:37.210 align:middle line:84% And the last key equation here is the capital accumulation 00:36:37.210 --> 00:36:37.760 align:middle line:90% equation. 00:36:37.760 --> 00:36:41.500 align:middle line:84% The capital accumulation equation says that capital, 00:36:41.500 --> 00:36:46.570 align:middle line:84% T plus 1, is equal to capital today, minus the depreciation-- 00:36:46.570 --> 00:36:50.050 align:middle line:84% some fraction of the machines break down in every period-- 00:36:50.050 --> 00:36:53.410 align:middle line:84% but plus the new investment, plus It. 00:36:53.410 --> 00:36:57.150 align:middle line:84% And we rewrote things and replaced the saving function 00:36:57.150 --> 00:36:57.650 align:middle line:90% and so on. 00:36:57.650 --> 00:37:01.030 align:middle line:84% And we end up in an expression like this 00:37:01.030 --> 00:37:05.050 align:middle line:84% that says capital per person here grows with investment, 00:37:05.050 --> 00:37:07.510 align:middle line:84% which is funded by savings, which is an increasing 00:37:07.510 --> 00:37:09.700 align:middle line:84% function of output, which in turn is an increasing 00:37:09.700 --> 00:37:12.400 align:middle line:84% function of capital per person, minus 00:37:12.400 --> 00:37:14.110 align:middle line:90% whatever is the depreciation. 00:37:14.110 --> 00:37:18.460 align:middle line:84% And what we did in the start diagram in the other model 00:37:18.460 --> 00:37:22.040 align:middle line:84% is, we plot this function and that function. 00:37:22.040 --> 00:37:25.360 align:middle line:84% We know that the steady state is when these two things are equal. 00:37:25.360 --> 00:37:30.800 align:middle line:84% That pins down the k star over, k over n star of this economy. 00:37:30.800 --> 00:37:33.580 align:middle line:84% But we also know that to the left 00:37:33.580 --> 00:37:38.760 align:middle line:84% of that point in capital space, this term is greater than that. 00:37:38.760 --> 00:37:41.960 align:middle line:84% And therefore, the capital stock is rising. 00:37:41.960 --> 00:37:45.950 align:middle line:84% To the right we get that this term is greater than that. 00:37:45.950 --> 00:37:48.420 align:middle line:84% And therefore the capital stock is falling. 00:37:48.420 --> 00:37:50.730 align:middle line:84% And that's what we have in this diagram. 00:37:50.730 --> 00:37:52.860 align:middle line:84% So that's the steady state of the economy, 00:37:52.860 --> 00:37:56.960 align:middle line:84% when the depreciation per worker, which is the required-- 00:37:56.960 --> 00:37:58.640 align:middle line:84% The minimum investment you need to keep 00:37:58.640 --> 00:38:00.440 align:middle line:84% the capital stock constant is whatever 00:38:00.440 --> 00:38:03.080 align:middle line:90% is depreciation per worker. 00:38:03.080 --> 00:38:06.137 align:middle line:84% Anything else you do, it will grow the stock of capital. 00:38:06.137 --> 00:38:07.970 align:middle line:84% Anything less you do, you're not maintaining 00:38:07.970 --> 00:38:09.178 align:middle line:90% enough of your capital stock. 00:38:09.178 --> 00:38:09.920 align:middle line:90% It's declining. 00:38:09.920 --> 00:38:11.690 align:middle line:84% And that's exactly what happens here. 00:38:11.690 --> 00:38:13.260 align:middle line:90% That's the steady state. 00:38:13.260 --> 00:38:17.930 align:middle line:84% If your capital below that, then a capital stock 00:38:17.930 --> 00:38:20.570 align:middle line:84% will be rising because you have lots of savings and therefore 00:38:20.570 --> 00:38:22.778 align:middle line:84% lots of investment relative to what you need in order 00:38:22.778 --> 00:38:26.060 align:middle line:84% to maintain the stock, the small stock of capital you have, 00:38:26.060 --> 00:38:29.660 align:middle line:90% until you reach steady state. 00:38:29.660 --> 00:38:36.290 align:middle line:84% And this model alone can explain really 00:38:36.290 --> 00:38:40.790 align:middle line:84% the pattern we have that the poorest economies tended 00:38:40.790 --> 00:38:42.990 align:middle line:84% to grow faster than the richer economies. 00:38:42.990 --> 00:38:45.110 align:middle line:84% If you think of poorer economies as economies 00:38:45.110 --> 00:38:49.940 align:middle line:84% that are otherwise similar, but that have a low stock of capital 00:38:49.940 --> 00:38:51.920 align:middle line:84% to start with, well, those economies 00:38:51.920 --> 00:38:53.960 align:middle line:84% are going to be to the left of the steady state, 00:38:53.960 --> 00:38:55.418 align:middle line:84% and therefore they're going to tend 00:38:55.418 --> 00:38:57.800 align:middle line:84% to grow at whatever is the steady state rate of growth, 00:38:57.800 --> 00:38:59.700 align:middle line:90% plus this catching up growth. 00:38:59.700 --> 00:39:03.760 align:middle line:84% And so this is a very powerful little model. 00:39:03.760 --> 00:39:08.060 align:middle line:84% It can explain a lot of those convergence, the convergence 00:39:08.060 --> 00:39:09.550 align:middle line:90% that we saw in the data. 00:39:09.550 --> 00:39:12.780 align:middle line:90% 00:39:12.780 --> 00:39:15.270 align:middle line:90% Do you understand this? 00:39:15.270 --> 00:39:16.390 align:middle line:90% This is important. 00:39:16.390 --> 00:39:21.580 align:middle line:90% 00:39:21.580 --> 00:39:22.830 align:middle line:90% Then we did some experiments. 00:39:22.830 --> 00:39:24.340 align:middle line:84% What happens if you increase the savings rate? 00:39:24.340 --> 00:39:26.570 align:middle line:84% At the time when Solow was writing this model, 00:39:26.570 --> 00:39:29.180 align:middle line:84% many people say that what was behind growth was saving. 00:39:29.180 --> 00:39:32.740 align:middle line:84% Well, in this model we show that, indeed, it 00:39:32.740 --> 00:39:34.320 align:middle line:90% is a saving rate rises. 00:39:34.320 --> 00:39:36.855 align:middle line:84% Then at any given level of capital-- 00:39:36.855 --> 00:39:38.480 align:middle line:84% suppose that was the oldest save state. 00:39:38.480 --> 00:39:40.600 align:middle line:84% If now the saving rate rises, that 00:39:40.600 --> 00:39:45.490 align:middle line:84% will increase investment above what 00:39:45.490 --> 00:39:48.190 align:middle line:84% you need to maintain that level of the stock of capital. 00:39:48.190 --> 00:39:50.860 align:middle line:84% So the stock of capital is going to start growing. 00:39:50.860 --> 00:39:53.170 align:middle line:84% When that happens, output per capita 00:39:53.170 --> 00:39:55.600 align:middle line:84% will grow faster than in a steady state 00:39:55.600 --> 00:39:57.880 align:middle line:84% because you're going to be going from here to there. 00:39:57.880 --> 00:40:00.912 align:middle line:84% But eventually you'll converge to the same old rate of growth. 00:40:00.912 --> 00:40:02.620 align:middle line:84% So the point here is that the saving rate 00:40:02.620 --> 00:40:06.040 align:middle line:84% can not change, per se, does not change the rate of growth 00:40:06.040 --> 00:40:06.730 align:middle line:90% in the long run. 00:40:06.730 --> 00:40:08.690 align:middle line:84% But it gives you transitional growth. 00:40:08.690 --> 00:40:12.760 align:middle line:84% And a lot of the Asian miracle of the very fast rates of growth 00:40:12.760 --> 00:40:16.480 align:middle line:84% from the '60s and '70s and '80s has to do with this kind 00:40:16.480 --> 00:40:19.720 align:middle line:84% of thing-- very sudden increase in saving rates plus other 00:40:19.720 --> 00:40:21.320 align:middle line:90% institutional changes and so on. 00:40:21.320 --> 00:40:23.690 align:middle line:84% But high increase in the savings rate 00:40:23.690 --> 00:40:27.410 align:middle line:90% also led to very fast growth. 00:40:27.410 --> 00:40:29.730 align:middle line:84% Again, this little model can explain a lot, as well. 00:40:29.730 --> 00:40:32.940 align:middle line:84% It can explain when you see those growth miracles, 00:40:32.940 --> 00:40:34.590 align:middle line:90% often it's associated to-- 00:40:34.590 --> 00:40:38.480 align:middle line:84% for some reason, it varies a lot across different scenarios-- 00:40:38.480 --> 00:40:42.460 align:middle line:84% the saving rate went up quite a bit. 00:40:42.460 --> 00:40:45.430 align:middle line:84% But point is, that gives you very fast growth 00:40:45.430 --> 00:40:47.355 align:middle line:84% in the short term, but eventually peters out. 00:40:47.355 --> 00:40:51.220 align:middle line:90% 00:40:51.220 --> 00:40:55.900 align:middle line:84% The next thing-- All that we did for a fixed population and said, 00:40:55.900 --> 00:40:58.905 align:middle line:84% well, suppose now the population is growing. 00:40:58.905 --> 00:41:01.510 align:middle line:90% 00:41:01.510 --> 00:41:03.160 align:middle line:84% I said the diagram we had before would 00:41:03.160 --> 00:41:05.743 align:middle line:84% be very unpleasant because all these curves would be shifting. 00:41:05.743 --> 00:41:08.440 align:middle line:84% So what we need to do is divide not by a constant. 00:41:08.440 --> 00:41:10.870 align:middle line:84% We need to divide by whatever is the population 00:41:10.870 --> 00:41:12.200 align:middle line:90% at that point in time. 00:41:12.200 --> 00:41:14.230 align:middle line:84% And that will give us the same diagram 00:41:14.230 --> 00:41:16.050 align:middle line:90% we had with one little twist. 00:41:16.050 --> 00:41:18.070 align:middle line:84% So I went through a little algebra 00:41:18.070 --> 00:41:22.600 align:middle line:84% here to arrive to a capital accumulation equation, capital 00:41:22.600 --> 00:41:26.140 align:middle line:84% per person equation, an equation for the change 00:41:26.140 --> 00:41:29.690 align:middle line:84% in the capital per person, which is very similar to what we had. 00:41:29.690 --> 00:41:33.550 align:middle line:84% The only difference is that the required capital, investment 00:41:33.550 --> 00:41:36.700 align:middle line:84% required to maintain the stock of capital per person, 00:41:36.700 --> 00:41:40.560 align:middle line:90% has an extra term here, gN. 00:41:40.560 --> 00:41:45.220 align:middle line:84% And I said that gN-- so let's think about this term. 00:41:45.220 --> 00:41:52.010 align:middle line:84% So delta, think of this as the required investment 00:41:52.010 --> 00:41:55.330 align:middle line:84% in order to maintain the stock of capital where it was. 00:41:55.330 --> 00:41:59.070 align:middle line:84% If this delta comes from the fact that, 00:41:59.070 --> 00:42:00.510 align:middle line:84% well, you have a stock of capital. 00:42:00.510 --> 00:42:01.685 align:middle line:90% You lose a fraction of that. 00:42:01.685 --> 00:42:03.810 align:middle line:84% Well, you need an investment equal to that fraction 00:42:03.810 --> 00:42:05.340 align:middle line:84% that you lost in order to maintain 00:42:05.340 --> 00:42:07.260 align:middle line:90% the stock of capital the same. 00:42:07.260 --> 00:42:10.560 align:middle line:90% That's clear. 00:42:10.560 --> 00:42:14.520 align:middle line:84% But that's not enough to maintain the capital per person 00:42:14.520 --> 00:42:19.290 align:middle line:84% constant if per person is rising, population is rising. 00:42:19.290 --> 00:42:23.550 align:middle line:84% Because even if you maintain the stock of capital constant, 00:42:23.550 --> 00:42:27.950 align:middle line:84% the denominator is rising at the rate of population growth. 00:42:27.950 --> 00:42:31.520 align:middle line:84% So in order to maintain the capital per person constant, 00:42:31.520 --> 00:42:35.010 align:middle line:84% you need to deal with the growth of the denominator, as well. 00:42:35.010 --> 00:42:37.010 align:middle line:84% And that means you need a little-- you need also 00:42:37.010 --> 00:42:39.770 align:middle line:84% investment to match the increase in population 00:42:39.770 --> 00:42:43.820 align:middle line:84% so they can keep the capital per person constant. 00:42:43.820 --> 00:42:46.230 align:middle line:90% So that's a modification. 00:42:46.230 --> 00:42:48.203 align:middle line:84% So in terms of our diagram, all that 00:42:48.203 --> 00:42:50.370 align:middle line:84% happened here, have technological progress, as well. 00:42:50.370 --> 00:42:51.690 align:middle line:90% Set it to 0 for now. 00:42:51.690 --> 00:42:53.930 align:middle line:84% All that happened relative to the previous diagram 00:42:53.930 --> 00:42:58.230 align:middle line:84% is that now I rotated this curve up, upward a little bit. 00:42:58.230 --> 00:43:04.210 align:middle line:84% But then you conduct analysis exactly the same way. 00:43:04.210 --> 00:43:05.850 align:middle line:84% The only thing that is different now 00:43:05.850 --> 00:43:08.270 align:middle line:84% is that in the previous model, we 00:43:08.270 --> 00:43:10.770 align:middle line:84% had that the rate of growth, the steady state rate of growth 00:43:10.770 --> 00:43:13.130 align:middle line:90% was equal to 0. 00:43:13.130 --> 00:43:15.710 align:middle line:84% And the steady state rate of growth of output per person 00:43:15.710 --> 00:43:16.985 align:middle line:90% was also equal to 0. 00:43:16.985 --> 00:43:18.110 align:middle line:90% Population was not growing. 00:43:18.110 --> 00:43:19.180 align:middle line:90% Output was not growing. 00:43:19.180 --> 00:43:22.100 align:middle line:90% The ratio wasn't growing either. 00:43:22.100 --> 00:43:25.610 align:middle line:84% Here it's still the case that in a steady state, 00:43:25.610 --> 00:43:29.160 align:middle line:84% output per person is not growing. 00:43:29.160 --> 00:43:32.740 align:middle line:84% But that also means since population 00:43:32.740 --> 00:43:35.320 align:middle line:84% is growing at the rate of gN or N, 00:43:35.320 --> 00:43:37.690 align:middle line:84% I don't know how we call it here, 00:43:37.690 --> 00:43:42.190 align:middle line:84% that means output must be also growing at the rate gN. 00:43:42.190 --> 00:43:46.180 align:middle line:84% That's what will keep output per person not growing. 00:43:46.180 --> 00:43:52.780 align:middle line:84% So for the output itself, a very important factor in growth 00:43:52.780 --> 00:43:54.170 align:middle line:90% is population growth. 00:43:54.170 --> 00:43:57.520 align:middle line:84% And if you look at rates of growth in general in the world, 00:43:57.520 --> 00:44:00.010 align:middle line:84% certainly in the developed world, 00:44:00.010 --> 00:44:03.040 align:middle line:84% they are falling for a variety of reasons. 00:44:03.040 --> 00:44:07.290 align:middle line:84% And one of them is because population growth is falling. 00:44:07.290 --> 00:44:10.680 align:middle line:84% But per person, that doesn't make a difference. 00:44:10.680 --> 00:44:13.710 align:middle line:84% But for the level of the rate of growth, it does. 00:44:13.710 --> 00:44:16.770 align:middle line:84% And then we added technological progress, 00:44:16.770 --> 00:44:21.460 align:middle line:84% which we modeled as effective as enhancing labor. 00:44:21.460 --> 00:44:26.920 align:middle line:84% So having a better technology means is you had more workers. 00:44:26.920 --> 00:44:28.530 align:middle line:84% So for any given level of workers, 00:44:28.530 --> 00:44:30.360 align:middle line:84% having a better technology, we modeled it 00:44:30.360 --> 00:44:32.820 align:middle line:90% as having more workers. 00:44:32.820 --> 00:44:35.380 align:middle line:84% And you can model it exactly that way. 00:44:35.380 --> 00:44:37.890 align:middle line:84% You can use exactly the same diagram we had before. 00:44:37.890 --> 00:44:40.273 align:middle line:84% But now we divide our scaling factor 00:44:40.273 --> 00:44:41.940 align:middle line:84% rather than being 1 over the population, 00:44:41.940 --> 00:44:44.250 align:middle line:84% it's going to be 1 over effective population, 00:44:44.250 --> 00:44:46.440 align:middle line:90% effective workers, 1 over AN. 00:44:46.440 --> 00:44:48.700 align:middle line:84% And you conduct exactly the same analysis. 00:44:48.700 --> 00:44:52.230 align:middle line:84% You do exactly the same approximations I did before. 00:44:52.230 --> 00:45:00.365 align:middle line:84% But the difference now is that here rather than have gN, 00:45:00.365 --> 00:45:02.640 align:middle line:90% you have AN. 00:45:02.640 --> 00:45:03.900 align:middle line:90% Why is that? 00:45:03.900 --> 00:45:06.810 align:middle line:84% Well, because if I want to maintain 00:45:06.810 --> 00:45:10.560 align:middle line:84% the capital per effective worker constant, 00:45:10.560 --> 00:45:13.440 align:middle line:84% then I need to first make up for the depreciation 00:45:13.440 --> 00:45:14.590 align:middle line:90% of the stock of capital. 00:45:14.590 --> 00:45:17.900 align:middle line:84% Thus, I have to stabilize the numerator. 00:45:17.900 --> 00:45:20.508 align:middle line:84% But then I have to take into account that the denominator is 00:45:20.508 --> 00:45:22.800 align:middle line:84% growing for two reasons-- because population is growing 00:45:22.800 --> 00:45:24.460 align:middle line:84% and because technology is growing. 00:45:24.460 --> 00:45:26.310 align:middle line:84% And in order to maintain the ratio constant, 00:45:26.310 --> 00:45:29.370 align:middle line:84% I'm going to have to add investment so to maintain 00:45:29.370 --> 00:45:30.503 align:middle line:90% that ratio constant. 00:45:30.503 --> 00:45:31.420 align:middle line:90% And that's the reason. 00:45:31.420 --> 00:45:37.230 align:middle line:84% Now we have this line here, rotates even further. 00:45:37.230 --> 00:45:39.030 align:middle line:84% And we get delta plus gA plus gN. 00:45:39.030 --> 00:45:42.967 align:middle line:90% 00:45:42.967 --> 00:45:44.550 align:middle line:84% And you should play with these things. 00:45:44.550 --> 00:45:46.550 align:middle line:84% What happened in this diagram if I 00:45:46.550 --> 00:45:48.990 align:middle line:90% increase gA or stuff like that. 00:45:48.990 --> 00:45:52.360 align:middle line:90% 00:45:52.360 --> 00:45:55.090 align:middle line:84% And notice that here, now still you have a steady state. 00:45:55.090 --> 00:45:56.590 align:middle line:84% But it's a steady state in the space 00:45:56.590 --> 00:45:59.650 align:middle line:84% of output per effective worker and capital 00:45:59.650 --> 00:46:01.060 align:middle line:90% per effective worker. 00:46:01.060 --> 00:46:04.930 align:middle line:84% That means, for example, that these quantities are not 00:46:04.930 --> 00:46:06.610 align:middle line:90% growing in the steady state. 00:46:06.610 --> 00:46:10.510 align:middle line:84% But output will be growing at which rate in the steady state? 00:46:10.510 --> 00:46:13.770 align:middle line:84% In any state, state here, what is the rate of growth of output? 00:46:13.770 --> 00:46:24.920 align:middle line:90% 00:46:24.920 --> 00:46:29.430 align:middle line:84% If output over AN is constant in the steady state, 00:46:29.430 --> 00:46:31.110 align:middle line:90% how can that happen? 00:46:31.110 --> 00:46:32.920 align:middle line:84% Output has to be growing at which rate? 00:46:32.920 --> 00:46:35.320 align:middle line:84% At the same rate as the denominator. 00:46:35.320 --> 00:46:37.470 align:middle line:90% So it's gA plus gN. 00:46:37.470 --> 00:46:41.890 align:middle line:90% 00:46:41.890 --> 00:46:43.100 align:middle line:90% That's a trickier question. 00:46:43.100 --> 00:46:47.290 align:middle line:84% What happens to output per person in this steady state? 00:46:47.290 --> 00:46:50.085 align:middle line:84% What rate is it growing at, output per person? 00:46:50.085 --> 00:46:53.310 align:middle line:90% 00:46:53.310 --> 00:46:55.486 align:middle line:84% Sorry, somebody said the right thing. 00:46:55.486 --> 00:46:57.790 align:middle line:90% gA, exactly. 00:46:57.790 --> 00:46:59.710 align:middle line:84% I want to keep this ratio constant. 00:46:59.710 --> 00:47:02.190 align:middle line:84% I'm asking the question of which space does this need 00:47:02.190 --> 00:47:04.500 align:middle line:84% to rise in order to maintain this constant, while 00:47:04.500 --> 00:47:06.550 align:middle line:84% at the same rate as A is growing. 00:47:06.550 --> 00:47:07.050 align:middle line:90% Good. 00:47:07.050 --> 00:47:09.910 align:middle line:90% 00:47:09.910 --> 00:47:12.130 align:middle line:84% Here we also did, we ask the question, well, 00:47:12.130 --> 00:47:14.110 align:middle line:84% could it be that here, if we change the saving 00:47:14.110 --> 00:47:16.115 align:middle line:84% rate, we get some extra kick in the long run? 00:47:16.115 --> 00:47:18.490 align:middle line:84% And the answer is no for the same logic as we had before. 00:47:18.490 --> 00:47:20.410 align:middle line:84% You're going to get transitional growth. 00:47:20.410 --> 00:47:23.237 align:middle line:84% But eventually you're going to convert to a steady state. 00:47:23.237 --> 00:47:24.820 align:middle line:84% And the rate of growth in the long run 00:47:24.820 --> 00:47:26.960 align:middle line:84% is not going to be a function of the saving rate. 00:47:26.960 --> 00:47:29.050 align:middle line:84% It's going to be equal to gA plus gN. 00:47:29.050 --> 00:47:42.330 align:middle line:90% 00:47:42.330 --> 00:47:43.490 align:middle line:90% I do here. 00:47:43.490 --> 00:47:45.560 align:middle line:84% Measuring technological progress, blah, blah. 00:47:45.560 --> 00:47:47.210 align:middle line:90% I told you the story of China. 00:47:47.210 --> 00:47:47.710 align:middle line:90% Good. 00:47:47.710 --> 00:47:52.620 align:middle line:90% 00:47:52.620 --> 00:47:54.120 align:middle line:90% We run out of time. 00:47:54.120 --> 00:47:59.720 align:middle line:84% So let me just say the last thing that I want to say. 00:47:59.720 --> 00:48:02.070 align:middle line:84% So the last thing we discussed, you 00:48:02.070 --> 00:48:04.890 align:middle line:84% say, well, what happen if you add education to this 00:48:04.890 --> 00:48:06.262 align:middle line:90% and try to-- 00:48:06.262 --> 00:48:09.630 align:middle line:90% [INAUDIBLE] 00:48:09.630 --> 00:48:11.220 align:middle line:84% I expanded the model a little bit. 00:48:11.220 --> 00:48:14.320 align:middle line:84% I had education and said, does it change conclusions a lot? 00:48:14.320 --> 00:48:15.940 align:middle line:90% I said, no, not really. 00:48:15.940 --> 00:48:18.630 align:middle line:84% I mean, it doesn't change the conclusion with respect 00:48:18.630 --> 00:48:19.950 align:middle line:90% to the long run. 00:48:19.950 --> 00:48:22.570 align:middle line:84% It affects the level of output per capita 00:48:22.570 --> 00:48:23.910 align:middle line:90% if you have more education. 00:48:23.910 --> 00:48:26.950 align:middle line:84% But you won't achieve effective rate of growth in the long run. 00:48:26.950 --> 00:48:33.030 align:middle line:84% And the last point I made is that, look, in this model, 00:48:33.030 --> 00:48:36.947 align:middle line:84% if you expand it and you try to assume that technology is 00:48:36.947 --> 00:48:39.030 align:middle line:84% the same and the rate of technological progress is 00:48:39.030 --> 00:48:41.970 align:middle line:84% the same across the world, you stick those parameters 00:48:41.970 --> 00:48:43.650 align:middle line:84% in the model-- population growth, 00:48:43.650 --> 00:48:45.720 align:middle line:84% the education levels and all that-- 00:48:45.720 --> 00:48:48.400 align:middle line:84% then you don't explain the amount of inequality 00:48:48.400 --> 00:48:49.390 align:middle line:90% we see in the world. 00:48:49.390 --> 00:48:51.190 align:middle line:84% The world will look a lot flatter 00:48:51.190 --> 00:48:57.390 align:middle line:84% if it was just a differences in population growth, deprivation, 00:48:57.390 --> 00:48:59.400 align:middle line:84% education levels, and things like that, 00:48:59.400 --> 00:49:01.090 align:middle line:90% but with the same technology. 00:49:01.090 --> 00:49:03.810 align:middle line:90% So if you want to account for-- 00:49:03.810 --> 00:49:06.300 align:middle line:84% so this model, the model essentially 00:49:06.300 --> 00:49:10.000 align:middle line:84% doesn't produce enough inequality in the world. 00:49:10.000 --> 00:49:11.490 align:middle line:84% You need to add something else that 00:49:11.490 --> 00:49:14.460 align:middle line:84% explains that we have some countries in Africa that are not 00:49:14.460 --> 00:49:16.960 align:middle line:84% growing or growing at a very low level of rate. 00:49:16.960 --> 00:49:20.070 align:middle line:84% And we said that something else technology, for whatever reason, 00:49:20.070 --> 00:49:24.150 align:middle line:84% it happens that there is a pocket of countries that 00:49:24.150 --> 00:49:27.090 align:middle line:84% seem to have a lower, sort of permanently lower 00:49:27.090 --> 00:49:31.030 align:middle line:84% level of technology at both level and growth rate. 00:49:31.030 --> 00:49:34.170 align:middle line:84% And that's what explains sort of a subset of countries that 00:49:34.170 --> 00:49:35.700 align:middle line:84% are sort of stacked, that are not 00:49:35.700 --> 00:49:37.625 align:middle line:84% consistent with this convergence type thing. 00:49:37.625 --> 00:49:40.000 align:middle line:84% So that's the reason it's called conditional convergence. 00:49:40.000 --> 00:49:42.570 align:middle line:84% Those countries themselves are converging to something. 00:49:42.570 --> 00:49:44.670 align:middle line:84% But they're converging to something much lower 00:49:44.670 --> 00:49:46.650 align:middle line:84% and with much lower rate of growth 00:49:46.650 --> 00:49:48.430 align:middle line:84% than most of the rest of the world. 00:49:48.430 --> 00:49:53.610 align:middle line:84% But the final lesson is for the average country, on average, 00:49:53.610 --> 00:49:56.310 align:middle line:84% it's clear that poorer countries grow faster 00:49:56.310 --> 00:49:57.610 align:middle line:90% than richer countries. 00:49:57.610 --> 00:49:59.375 align:middle line:90% That's a dominant force. 00:49:59.375 --> 00:50:00.750 align:middle line:84% But you need a little more if you 00:50:00.750 --> 00:50:03.420 align:middle line:84% want to explain certain pockets of the world. 00:50:03.420 --> 00:50:04.970 align:middle line:90% OK. 00:50:04.970 --> 00:50:23.000 align:middle line:90%