WEBVTT 00:00:00.000 --> 00:00:01.984 align:middle line:90% [SQUEAKING] 00:00:01.984 --> 00:00:03.968 align:middle line:90% [RUSTLING] 00:00:03.968 --> 00:00:05.952 align:middle line:90% [CLICKING] 00:00:05.952 --> 00:00:11.910 align:middle line:90% 00:00:11.910 --> 00:00:13.530 align:middle line:84% ROBERT TOWNSEND: Welcome, everyone. 00:00:13.530 --> 00:00:14.850 align:middle line:90% Try to start on time. 00:00:14.850 --> 00:00:18.022 align:middle line:84% Are there any questions from last time? 00:00:18.022 --> 00:00:22.200 align:middle line:84% OK, we'll take a look at the class syllabus. 00:00:22.200 --> 00:00:25.710 align:middle line:84% We did a theory lecture on Walrasian equilibria 00:00:25.710 --> 00:00:32.009 align:middle line:84% and 2x2x2 trade models in theory, last class. 00:00:32.009 --> 00:00:33.780 align:middle line:84% Today, we're going to begin the first 00:00:33.780 --> 00:00:36.420 align:middle line:90% of two applications of that. 00:00:36.420 --> 00:00:40.290 align:middle line:84% You can see, in fact, on the reading list the papers 00:00:40.290 --> 00:00:44.580 align:middle line:84% for today, which includes this thing with my Thai colleague, 00:00:44.580 --> 00:00:47.640 align:middle line:84% Archawa, which is taking the theory model back 00:00:47.640 --> 00:00:49.830 align:middle line:90% to the Thai applications. 00:00:49.830 --> 00:00:52.560 align:middle line:84% And there's an additional stared reading here, 00:00:52.560 --> 00:00:55.860 align:middle line:84% Dawkins, which is calibration, which 00:00:55.860 --> 00:00:59.730 align:middle line:84% we have done before with the medieval villages and dynamics. 00:00:59.730 --> 00:01:02.310 align:middle line:84% And today, we're going to do it again. 00:01:02.310 --> 00:01:05.319 align:middle line:84% But that start reading does not appear anywhere, 00:01:05.319 --> 00:01:07.200 align:middle line:90% particularly in the lectures. 00:01:07.200 --> 00:01:10.230 align:middle line:84% But still, there aren't a lot of new theorems 00:01:10.230 --> 00:01:13.150 align:middle line:84% and stuff in the lecture today, mostly concepts. 00:01:13.150 --> 00:01:17.310 align:middle line:84% So I think you can get some time to look at the Dawkins starred 00:01:17.310 --> 00:01:18.330 align:middle line:90% article. 00:01:18.330 --> 00:01:21.610 align:middle line:90% And on the class schedule-- 00:01:21.610 --> 00:01:25.830 align:middle line:84% so after the trade lecture application today and the one 00:01:25.830 --> 00:01:28.020 align:middle line:90% next Thursday. 00:01:28.020 --> 00:01:31.140 align:middle line:84% And then the following week, on Tuesday, 00:01:31.140 --> 00:01:36.120 align:middle line:84% I'll conduct with you a review session of the material covered 00:01:36.120 --> 00:01:38.680 align:middle line:90% in the last 1/3 of the class. 00:01:38.680 --> 00:01:42.390 align:middle line:84% So that's the way the schedule looks. 00:01:42.390 --> 00:01:45.580 align:middle line:84% Again, let me ask if there are questions. 00:01:45.580 --> 00:01:48.420 align:middle line:84% Otherwise, I'm going to turn to the study guide 00:01:48.420 --> 00:01:54.180 align:middle line:84% and, in particular, the material from last time lecture 13. 00:01:54.180 --> 00:01:58.410 align:middle line:84% So the first review question was, in addition 00:01:58.410 --> 00:02:02.100 align:middle line:84% to reviewing what it takes in general equilibrium terms 00:02:02.100 --> 00:02:06.020 align:middle line:84% to describe an economy, what additional features do 00:02:06.020 --> 00:02:11.750 align:middle line:84% we need to put in there to describe a private ownership 00:02:11.750 --> 00:02:13.810 align:middle line:90% economy? 00:02:13.810 --> 00:02:17.250 align:middle line:90% And let me see if-- 00:02:17.250 --> 00:02:19.245 align:middle line:84% Dixon, do you want to take a stab at that? 00:02:19.245 --> 00:02:22.140 align:middle line:90% 00:02:22.140 --> 00:02:26.420 align:middle line:84% AUDIENCE: Yeah, so for general equilibrium [INAUDIBLE] 00:02:26.420 --> 00:02:30.800 align:middle line:84% we need like the consumers' like preferences and consumption 00:02:30.800 --> 00:02:36.680 align:middle line:84% set, the firm technology, which is the production set, and then 00:02:36.680 --> 00:02:40.580 align:middle line:84% economy resources, which is endowment. 00:02:40.580 --> 00:02:45.150 align:middle line:84% But additionally, for the private ownership, 00:02:45.150 --> 00:02:49.920 align:middle line:84% each consumer has their own endowment. 00:02:49.920 --> 00:02:55.140 align:middle line:84% And also, each consumer has a complete ownership of the firm. 00:02:55.140 --> 00:02:59.577 align:middle line:84% So the firm is made out of consumer ownership. 00:02:59.577 --> 00:03:00.660 align:middle line:90% ROBERT TOWNSEND: Good, OK. 00:03:00.660 --> 00:03:04.500 align:middle line:84% Yeah, each consumer has shares in firms. 00:03:04.500 --> 00:03:08.520 align:middle line:84% So those were the theta ij parameters. 00:03:08.520 --> 00:03:12.780 align:middle line:84% So if there are profits, as in decreasing returns to scale, 00:03:12.780 --> 00:03:17.500 align:middle line:84% those profits get distributed back to households. 00:03:17.500 --> 00:03:19.488 align:middle line:84% So they own the returns on the firm. 00:03:19.488 --> 00:03:20.530 align:middle line:90% So that's a great answer. 00:03:20.530 --> 00:03:21.280 align:middle line:90% Thank you so much. 00:03:21.280 --> 00:03:24.730 align:middle line:90% 00:03:24.730 --> 00:03:29.430 align:middle line:84% What is the potential difference between a Walrasian equilibrium 00:03:29.430 --> 00:03:35.000 align:middle line:84% and a price equilibrium with transfers? 00:03:35.000 --> 00:03:38.340 align:middle line:84% Maria Sofia, do you want to try that one? 00:03:38.340 --> 00:03:41.250 align:middle line:84% AUDIENCE: Yeah, so with the price equilibrium 00:03:41.250 --> 00:03:43.110 align:middle line:84% with transfers, we're looking to see 00:03:43.110 --> 00:03:48.270 align:middle line:84% if there exists some assignment of wealth levels 00:03:48.270 --> 00:03:51.790 align:middle line:84% such that the economy will satisfy a set of constraints. 00:03:51.790 --> 00:03:55.260 align:middle line:84% But in general, a Walrasian equilibrium 00:03:55.260 --> 00:04:00.000 align:middle line:84% is a type of price equilibrium with transfers, 00:04:00.000 --> 00:04:05.460 align:middle line:84% except we have the condition that the wealth is equal to-- 00:04:05.460 --> 00:04:07.470 align:middle line:84% I think it's the price times the wealth 00:04:07.470 --> 00:04:11.520 align:middle line:84% plus the sum of the profits over the shares 00:04:11.520 --> 00:04:15.010 align:middle line:84% that the consumer has in various companies. 00:04:15.010 --> 00:04:16.510 align:middle line:84% ROBERT TOWNSEND: Yes, that's right. 00:04:16.510 --> 00:04:20.089 align:middle line:84% So the Walrasian equilibrium works off 00:04:20.089 --> 00:04:22.230 align:middle line:84% of this private ownership economy. 00:04:22.230 --> 00:04:25.280 align:middle line:84% So we have the valuation of the household's endowment 00:04:25.280 --> 00:04:28.900 align:middle line:84% plus the valuation of the profits 00:04:28.900 --> 00:04:33.070 align:middle line:84% that the household has a claim to. 00:04:33.070 --> 00:04:37.780 align:middle line:90% And that is a wealth object. 00:04:37.780 --> 00:04:41.280 align:middle line:84% So technically, it looks like a price equilibrium 00:04:41.280 --> 00:04:44.550 align:middle line:84% with transfers, meaning that if we 00:04:44.550 --> 00:04:46.350 align:middle line:84% assign that particular wealth, we're 00:04:46.350 --> 00:04:48.480 align:middle line:90% going to have an equilibrium. 00:04:48.480 --> 00:04:52.800 align:middle line:84% The more general idea behind the price equilibrium 00:04:52.800 --> 00:04:57.570 align:middle line:84% with transfers, though, including the transfers part, 00:04:57.570 --> 00:05:01.560 align:middle line:84% is that the wealth assignments that we give to households 00:05:01.560 --> 00:05:05.850 align:middle line:84% may be greater or less than the valuation of the wealth they 00:05:05.850 --> 00:05:12.340 align:middle line:84% would have if we had restricted ourselves to private ownership. 00:05:12.340 --> 00:05:16.350 align:middle line:84% So what I didn't, I guess, emphasize very much is 00:05:16.350 --> 00:05:19.410 align:middle line:84% there are going to be two welfare theorems 00:05:19.410 --> 00:05:21.750 align:middle line:90% that we will study in detail. 00:05:21.750 --> 00:05:27.880 align:middle line:84% One is that the Walrasian equilibrium is pareto optimal. 00:05:27.880 --> 00:05:31.970 align:middle line:84% And for that part, we need private ownership only. 00:05:31.970 --> 00:05:35.990 align:middle line:84% And the second is going to be that any pareto optimal 00:05:35.990 --> 00:05:41.270 align:middle line:84% allocation can be achieved with the suitable redistribution 00:05:41.270 --> 00:05:42.248 align:middle line:90% of wealth. 00:05:42.248 --> 00:05:43.790 align:middle line:84% And that's the part where we're going 00:05:43.790 --> 00:05:47.600 align:middle line:84% to need this more general concept of price equilibrium 00:05:47.600 --> 00:05:48.650 align:middle line:90% with transfers. 00:05:48.650 --> 00:05:51.580 align:middle line:90% 00:05:51.580 --> 00:05:58.830 align:middle line:84% The ordering of the class was to introduce this Walrasian 00:05:58.830 --> 00:06:04.600 align:middle line:84% equilibrium notion, not dwell on the optimality of it, 00:06:04.600 --> 00:06:09.030 align:middle line:84% but rather get into some applications and then. 00:06:09.030 --> 00:06:11.910 align:middle line:84% In a couple of weeks, we'll be back to those welfare theorems 00:06:11.910 --> 00:06:15.350 align:middle line:84% and when they are true and when they can fail. 00:06:15.350 --> 00:06:18.910 align:middle line:84% So that's the ordering of the class. 00:06:18.910 --> 00:06:25.280 align:middle line:90% And we got into the 2x2x2. 00:06:25.280 --> 00:06:26.750 align:middle line:84% All of this stuff at the beginning 00:06:26.750 --> 00:06:32.010 align:middle line:84% is kind of a review of the notation, where 00:06:32.010 --> 00:06:38.910 align:middle line:84% we have minimizing the cost of achieving a unit output, 00:06:38.910 --> 00:06:41.610 align:middle line:84% given the vector of factor prices w. 00:06:41.610 --> 00:06:45.770 align:middle line:84% And we had the cost-minimizing input values. 00:06:45.770 --> 00:06:48.430 align:middle line:90% 00:06:48.430 --> 00:06:51.900 align:middle line:84% And at the end of all that, I finally got around to asking, 00:06:51.900 --> 00:06:54.540 align:middle line:84% define what it means for the production of good 1 00:06:54.540 --> 00:06:59.430 align:middle line:84% to be relatively more intensive in factor 1 00:06:59.430 --> 00:07:05.180 align:middle line:90% than the production of good 2. 00:07:05.180 --> 00:07:11.560 align:middle line:84% So [INAUDIBLE] see if we can get some-- 00:07:11.560 --> 00:07:15.170 align:middle line:84% how about a volunteer for that one? 00:07:15.170 --> 00:07:17.932 align:middle line:90% AUDIENCE: I can take this one. 00:07:17.932 --> 00:07:26.510 align:middle line:84% Yeah, so this means that for producing a unit of good 1 00:07:26.510 --> 00:07:33.350 align:middle line:84% versus a unit of good 2, you need the ratio of factor 1 00:07:33.350 --> 00:07:36.260 align:middle line:84% to factor 2 that's used to produce one unit of good 1 00:07:36.260 --> 00:07:40.100 align:middle line:84% is greater than the ratio of factor 1 00:07:40.100 --> 00:07:43.160 align:middle line:84% to factor 2 in producing a unit of good 2. 00:07:43.160 --> 00:07:44.910 align:middle line:90% ROBERT TOWNSEND: Excellent. 00:07:44.910 --> 00:07:46.510 align:middle line:90% Perfect answer. 00:07:46.510 --> 00:07:49.560 align:middle line:84% And then implicit, let me just add a little bit 00:07:49.560 --> 00:07:55.400 align:middle line:84% editorial, which is, it's defined over the whole domain. 00:07:55.400 --> 00:08:00.420 align:middle line:84% So whatever those factor prices are, w1 in w2, 00:08:00.420 --> 00:08:05.170 align:middle line:84% it remains true that good one uses relatively more factor 00:08:05.170 --> 00:08:08.240 align:middle line:90% 1 than factor 2 and vice versa. 00:08:08.240 --> 00:08:10.210 align:middle line:84% So it's a statement about the whole-- 00:08:10.210 --> 00:08:13.990 align:middle line:84% and we used that in the analysis in class, 00:08:13.990 --> 00:08:15.910 align:middle line:84% when we were looking at the cost curves 00:08:15.910 --> 00:08:19.810 align:middle line:84% and what would happen if input prices moved. 00:08:19.810 --> 00:08:24.430 align:middle line:90% 00:08:24.430 --> 00:08:29.440 align:middle line:84% Can someone take a stab at the Stolper-Samuelson theorem 00:08:29.440 --> 00:08:34.375 align:middle line:84% for the determination of input prices as output prices change? 00:08:34.375 --> 00:08:37.539 align:middle line:90% 00:08:37.539 --> 00:08:38.679 align:middle line:90% AUDIENCE: I can. 00:08:38.679 --> 00:08:40.110 align:middle line:90% ROBERT TOWNSEND: OK. 00:08:40.110 --> 00:08:46.290 align:middle line:84% AUDIENCE: In this theorem, we assume 00:08:46.290 --> 00:08:51.030 align:middle line:84% that the model [INAUDIBLE] to modify [INAUDIBLE] then when 00:08:51.030 --> 00:08:55.590 align:middle line:84% the output price, like the price of good 1, 00:08:55.590 --> 00:09:03.660 align:middle line:84% increases, then the intensive factors in producing the good 1 00:09:03.660 --> 00:09:05.820 align:middle line:90% also increases. 00:09:05.820 --> 00:09:10.260 align:middle line:84% And another factor, factor 2, that 00:09:10.260 --> 00:09:13.320 align:middle line:84% was not the intensive factor [INAUDIBLE] 00:09:13.320 --> 00:09:16.060 align:middle line:84% the price of this factor world goes down. 00:09:16.060 --> 00:09:16.560 align:middle line:90% Yeah. 00:09:16.560 --> 00:09:17.935 align:middle line:84% ROBERT TOWNSEND: OK, that's good. 00:09:17.935 --> 00:09:19.380 align:middle line:90% Yup. 00:09:19.380 --> 00:09:21.330 align:middle line:84% So nominally, it's not a statement 00:09:21.330 --> 00:09:23.220 align:middle line:84% about international trade, although it 00:09:23.220 --> 00:09:26.510 align:middle line:90% is very useful in that context. 00:09:26.510 --> 00:09:31.650 align:middle line:84% We're just basically doing some comparative static exercises 00:09:31.650 --> 00:09:36.480 align:middle line:84% of what would happen if these exogenously given output 00:09:36.480 --> 00:09:37.650 align:middle line:90% prices are changing. 00:09:37.650 --> 00:09:39.375 align:middle line:84% What's going to happen to input prices? 00:09:39.375 --> 00:09:43.500 align:middle line:90% 00:09:43.500 --> 00:09:46.050 align:middle line:90% And then we did a graph. 00:09:46.050 --> 00:09:49.520 align:middle line:84% We used those cost curves to illustrate it. 00:09:49.520 --> 00:09:51.440 align:middle line:84% I flagged the last question, but I really 00:09:51.440 --> 00:09:53.240 align:middle line:90% don't want to ask that one. 00:09:53.240 --> 00:09:57.930 align:middle line:84% I want to ask something a little harder to answer. 00:09:57.930 --> 00:10:01.260 align:middle line:90% 00:10:01.260 --> 00:10:05.740 align:middle line:84% If we had two countries, and each country had 00:10:05.740 --> 00:10:09.120 align:middle line:84% a representative consumer, and there 00:10:09.120 --> 00:10:15.010 align:middle line:84% are two goods, each of which can be produced in both countries, 00:10:15.010 --> 00:10:16.810 align:middle line:84% and there were two factors that can 00:10:16.810 --> 00:10:21.060 align:middle line:84% be used in each good in each country, 00:10:21.060 --> 00:10:25.210 align:middle line:84% we've got a lot of equations and unknowns. 00:10:25.210 --> 00:10:28.000 align:middle line:84% We've got the consumers' consumption 00:10:28.000 --> 00:10:32.680 align:middle line:84% levels of the two goods, the input being 00:10:32.680 --> 00:10:35.980 align:middle line:84% used to produce outputs of the two goods, 00:10:35.980 --> 00:10:39.160 align:middle line:84% the outputs themselves of the two goods, 00:10:39.160 --> 00:10:44.750 align:middle line:84% and the prices of each of the two inputs in both countries, 00:10:44.750 --> 00:10:48.050 align:middle line:90% et cetera, et cetera. 00:10:48.050 --> 00:10:52.190 align:middle line:84% Can someone tell me how we attack this problem in class, 00:10:52.190 --> 00:10:55.530 align:middle line:90% in the sense of an algorithm? 00:10:55.530 --> 00:10:57.150 align:middle line:90% We broke it into pieces. 00:10:57.150 --> 00:11:00.200 align:middle line:90% 00:11:00.200 --> 00:11:03.785 align:middle line:84% Otherwise, it's a very high-dimensional problem. 00:11:03.785 --> 00:11:08.790 align:middle line:90% 00:11:08.790 --> 00:11:10.050 align:middle line:90% If you remember the ordering-- 00:11:10.050 --> 00:11:10.380 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:11:10.380 --> 00:11:12.240 align:middle line:84% ROBERT TOWNSEND: Yeah, go ahead, please. 00:11:12.240 --> 00:11:13.740 align:middle line:84% AUDIENCE: Yeah, I think that the way 00:11:13.740 --> 00:11:16.980 align:middle line:84% that we did this was kind of inductively, 00:11:16.980 --> 00:11:19.740 align:middle line:84% where I think that the way that we did this was, 00:11:19.740 --> 00:11:22.950 align:middle line:84% first, pretend, basically, that it's just 00:11:22.950 --> 00:11:24.720 align:middle line:90% like the original 2x2 model. 00:11:24.720 --> 00:11:27.970 align:middle line:84% And then we solve for the what's supposed to happen there. 00:11:27.970 --> 00:11:30.300 align:middle line:84% And then we built in the additional thing of-- 00:11:30.300 --> 00:11:33.990 align:middle line:84% I'm pretty sure we built in the different endowments, I think, 00:11:33.990 --> 00:11:37.920 align:middle line:84% was the difference in the 2x2x2 versus the 2x2. 00:11:37.920 --> 00:11:39.480 align:middle line:84% But we solve for the smaller model, 00:11:39.480 --> 00:11:42.550 align:middle line:84% and then we build it upwards to the bigger one. 00:11:42.550 --> 00:11:44.090 align:middle line:90% ROBERT TOWNSEND: That's true. 00:11:44.090 --> 00:11:45.790 align:middle line:84% And what, in particular, did we focus 00:11:45.790 --> 00:11:47.335 align:middle line:90% on first in the smaller model? 00:11:47.335 --> 00:11:50.250 align:middle line:90% 00:11:50.250 --> 00:11:53.550 align:middle line:84% AUDIENCE: I think the way that we did this was kind of the-- 00:11:53.550 --> 00:11:59.240 align:middle line:90% 00:11:59.240 --> 00:12:01.060 align:middle line:90% I don't think I remember this. 00:12:01.060 --> 00:12:04.370 align:middle line:84% ROBERT TOWNSEND: OK, anyone else want to take an attempt at it? 00:12:04.370 --> 00:12:09.980 align:middle line:84% This is a difficult question to pose and hard to answer, 00:12:09.980 --> 00:12:13.540 align:middle line:84% because it has to do with putting 00:12:13.540 --> 00:12:17.290 align:middle line:84% the pieces a lot of information in the lecture 00:12:17.290 --> 00:12:21.460 align:middle line:84% and taking a step backwards and thinking about it globally. 00:12:21.460 --> 00:12:28.450 align:middle line:90% 00:12:28.450 --> 00:12:32.890 align:middle line:84% OK, well, it starts with this Stolper-Samuelson theorem. 00:12:32.890 --> 00:12:35.950 align:middle line:84% So at this point, early on in the lecture, 00:12:35.950 --> 00:12:40.420 align:middle line:84% we were taking the prices of the goods p1 and p2, as given. 00:12:40.420 --> 00:12:42.130 align:middle line:84% We weren't trying to solve for that. 00:12:42.130 --> 00:12:45.970 align:middle line:84% I had not yet introduced consumers either. 00:12:45.970 --> 00:12:49.960 align:middle line:84% We didn't even attempt to solve for the quantities. 00:12:49.960 --> 00:12:55.060 align:middle line:84% We merely wanted to find equilibrium factor prices. 00:12:55.060 --> 00:13:00.100 align:middle line:90% So for a given country-- 00:13:00.100 --> 00:13:02.730 align:middle line:84% you could think of it that way, a given economy-- 00:13:02.730 --> 00:13:08.920 align:middle line:84% we had the minimized cost of producing one unit of good 1, 00:13:08.920 --> 00:13:14.170 align:middle line:84% the minimized cost of producing one unit of good 2. 00:13:14.170 --> 00:13:23.620 align:middle line:84% And that's in the space of input prices w1 and w2. 00:13:23.620 --> 00:13:28.840 align:middle line:84% So we're looking for an equilibrium. 00:13:28.840 --> 00:13:30.480 align:middle line:84% Therefore, the input prices should 00:13:30.480 --> 00:13:35.100 align:middle line:84% be the same regardless of whether it's good 1 or good 2. 00:13:35.100 --> 00:13:37.290 align:middle line:84% So we looked at where those curves crossed, 00:13:37.290 --> 00:13:43.020 align:middle line:84% and that's the only point in the diagram where we can both have 00:13:43.020 --> 00:13:50.160 align:middle line:84% common input prices of the two inputs and be on those curves, 00:13:50.160 --> 00:13:52.050 align:middle line:90% where the curves crossed. 00:13:52.050 --> 00:13:55.650 align:middle line:84% So that was the very, very first part. 00:13:55.650 --> 00:14:04.200 align:middle line:84% And then given those prices, we moved to a different space, 00:14:04.200 --> 00:14:06.300 align:middle line:90% which had isoquants. 00:14:06.300 --> 00:14:13.000 align:middle line:84% So there, we have the unit output or any output, 00:14:13.000 --> 00:14:18.900 align:middle line:84% really, described by an isoquant, a different isoquant 00:14:18.900 --> 00:14:25.810 align:middle line:84% for different quantities, and we had the inputs on the x-axis. 00:14:25.810 --> 00:14:31.300 align:middle line:84% So there, we solved for minimizing unit cost again, 00:14:31.300 --> 00:14:35.930 align:middle line:84% but what's on the x and y-axis would be the input quantities. 00:14:35.930 --> 00:14:42.870 align:middle line:84% And we know that has to align with the factor 00:14:42.870 --> 00:14:46.290 align:middle line:90% prices we already solved for. 00:14:46.290 --> 00:14:51.960 align:middle line:84% So there, we determine both the input quantities as well as 00:14:51.960 --> 00:14:58.150 align:middle line:84% the output quantities, where both firm 1 and firm 2 00:14:58.150 --> 00:15:01.030 align:middle line:90% could be in equilibrium. 00:15:01.030 --> 00:15:04.430 align:middle line:84% And then, finally, at the very end of the lecture, 00:15:04.430 --> 00:15:12.460 align:middle line:84% we zeroed out to consider two countries, not just one, each 00:15:12.460 --> 00:15:14.590 align:middle line:90% with a consumer. 00:15:14.590 --> 00:15:19.240 align:middle line:84% And those countries could be in autarky or potentially trading 00:15:19.240 --> 00:15:20.005 align:middle line:90% with one another. 00:15:20.005 --> 00:15:23.160 align:middle line:90% 00:15:23.160 --> 00:15:27.990 align:middle line:84% Both specifications would be legitimate general equilibrium 00:15:27.990 --> 00:15:30.060 align:middle line:90% environments, both autarky-- 00:15:30.060 --> 00:15:32.640 align:middle line:84% each country separately is one-- and then 00:15:32.640 --> 00:15:35.730 align:middle line:84% we traded another, except, obviously, 00:15:35.730 --> 00:15:38.190 align:middle line:84% with trade, consumption and production 00:15:38.190 --> 00:15:41.610 align:middle line:84% do not have to be the same, because the countries can 00:15:41.610 --> 00:15:44.430 align:middle line:84% export one thing and import another. 00:15:44.430 --> 00:15:47.070 align:middle line:84% Sorry, that's a rather long-winded summary. 00:15:47.070 --> 00:15:51.120 align:middle line:84% But again, it's not the easiest lecture, maybe 00:15:51.120 --> 00:15:55.060 align:middle line:84% potentially the hardest lecture in the whole series 00:15:55.060 --> 00:15:55.970 align:middle line:90% of lectures. 00:15:55.970 --> 00:15:58.750 align:middle line:84% So I thought I would take the time 00:15:58.750 --> 00:16:03.530 align:middle line:84% to elaborate a bit on what we were doing. 00:16:03.530 --> 00:16:08.880 align:middle line:90% OK, so that's the review. 00:16:08.880 --> 00:16:15.520 align:middle line:84% And we come to applying a very similar theory to actual data. 00:16:15.520 --> 00:16:18.270 align:middle line:84% So this is real and financial flows 00:16:18.270 --> 00:16:21.240 align:middle line:90% in Thailand and an outline. 00:16:21.240 --> 00:16:23.700 align:middle line:84% We're going to talk about villages again, 00:16:23.700 --> 00:16:27.480 align:middle line:84% but we're going to embed them in the local economy 00:16:27.480 --> 00:16:31.080 align:middle line:84% and also their relationship to the national economy, where 00:16:31.080 --> 00:16:33.540 align:middle line:84% we view the financial accounts, which 00:16:33.540 --> 00:16:38.580 align:middle line:84% you've seen once before when we did the Bitcoin part. 00:16:38.580 --> 00:16:40.140 align:middle line:90% We'll review them here-- 00:16:40.140 --> 00:16:43.200 align:middle line:84% balance sheets, income statement, and cash flow-- 00:16:43.200 --> 00:16:47.160 align:middle line:84% then more general background on these village economies 00:16:47.160 --> 00:16:49.440 align:middle line:90% and heterogeneity within them. 00:16:49.440 --> 00:16:54.090 align:middle line:84% Then we'll take the village economy, as an entire economy, 00:16:54.090 --> 00:16:59.370 align:middle line:84% and come up with GDP and village income, 00:16:59.370 --> 00:17:01.890 align:middle line:84% the way we come up with national income and the National 00:17:01.890 --> 00:17:03.930 align:middle line:90% Income and Product Accounts. 00:17:03.930 --> 00:17:07.300 align:middle line:84% And then, finally, as a small open economy, 00:17:07.300 --> 00:17:10.650 align:middle line:84% we can look at flow of funds, balance of payments, how open 00:17:10.650 --> 00:17:12.900 align:middle line:90% or how closed is the economy. 00:17:12.900 --> 00:17:15.099 align:middle line:84% And then we come to this calibration part, 00:17:15.099 --> 00:17:17.430 align:middle line:84% which I already highlighted, with respect 00:17:17.430 --> 00:17:20.430 align:middle line:84% to the other starred reading, Dawkins et al. 00:17:20.430 --> 00:17:22.349 align:middle line:84% What does it mean to be calibrating 00:17:22.349 --> 00:17:25.470 align:middle line:84% a model, which they often did in the trade context, 00:17:25.470 --> 00:17:27.329 align:middle line:90% as we do here? 00:17:27.329 --> 00:17:30.600 align:middle line:84% And we'll look at model-predicted data 00:17:30.600 --> 00:17:33.660 align:middle line:84% and compare it directly to actual data 00:17:33.660 --> 00:17:35.970 align:middle line:84% and then look at some counterfactuals, 00:17:35.970 --> 00:17:39.760 align:middle line:84% like imposing tariffs and frictions. 00:17:39.760 --> 00:17:43.050 align:middle line:84% So last time end today, you could think of this question 00:17:43.050 --> 00:17:47.340 align:middle line:84% as, free trade, who are the winners and losers, tariffs, 00:17:47.340 --> 00:17:50.730 align:middle line:84% who are the winners and losers, and not in the abstract, 00:17:50.730 --> 00:17:53.100 align:middle line:84% as in last time, but actually here, 00:17:53.100 --> 00:17:55.740 align:middle line:84% in the context of a calibrated model. 00:17:55.740 --> 00:18:00.510 align:middle line:84% You'll actually see the impact on individual people living 00:18:00.510 --> 00:18:01.980 align:middle line:90% in these villages. 00:18:01.980 --> 00:18:04.110 align:middle line:84% And then when we get to the US, we'll 00:18:04.110 --> 00:18:07.740 align:middle line:84% do it all again on Thursday, which is a different context, 00:18:07.740 --> 00:18:09.360 align:middle line:90% but the same kind of theory. 00:18:09.360 --> 00:18:11.820 align:middle line:90% OK, so you've seen this before. 00:18:11.820 --> 00:18:14.400 align:middle line:90% Let me highlight the time map. 00:18:14.400 --> 00:18:18.270 align:middle line:84% We have these two provinces, Chachoengsao and Lopburi 00:18:18.270 --> 00:18:20.850 align:middle line:84% in the central area of the country, 00:18:20.850 --> 00:18:23.850 align:middle line:84% near, if not right on top of, Bangkok. 00:18:23.850 --> 00:18:27.420 align:middle line:84% And we have the other two, Buriram and Sisaket 00:18:27.420 --> 00:18:31.920 align:middle line:84% in the Northeast, a much more agrarian area of the country. 00:18:31.920 --> 00:18:33.450 align:middle line:90% This is Sisaket. 00:18:33.450 --> 00:18:37.410 align:middle line:84% I muttered something about not finding these monthly starred 00:18:37.410 --> 00:18:39.570 align:middle line:84% villages, but I knew where they were. 00:18:39.570 --> 00:18:42.270 align:middle line:84% And now I realize it's still very hard to read. 00:18:42.270 --> 00:18:47.010 align:middle line:84% But these Cheerio circles correspond to baseline survey 00:18:47.010 --> 00:18:48.270 align:middle line:90% done in 1997. 00:18:48.270 --> 00:18:50.820 align:middle line:84% A subset of those villages were chosen 00:18:50.820 --> 00:18:54.420 align:middle line:84% from '98 onward, including the data that we're using. 00:18:54.420 --> 00:18:56.310 align:middle line:84% In which case, it's a monthly survey. 00:18:56.310 --> 00:18:58.830 align:middle line:84% So the Cheerio covers up the star 00:18:58.830 --> 00:19:01.780 align:middle line:84% and produces these green-like objects up here. 00:19:01.780 --> 00:19:02.920 align:middle line:90% So that's the cluster. 00:19:02.920 --> 00:19:05.330 align:middle line:84% So there are four villages in Sisaket 00:19:05.330 --> 00:19:07.650 align:middle line:84% that constitute the monthly data. 00:19:07.650 --> 00:19:10.830 align:middle line:84% And likewise, there are going to be four each in the other three 00:19:10.830 --> 00:19:12.060 align:middle line:90% provinces. 00:19:12.060 --> 00:19:18.870 align:middle line:84% OK so let's take Lopburi over here, and take a wage earner, 00:19:18.870 --> 00:19:24.070 align:middle line:84% and look at this actual household's income statement. 00:19:24.070 --> 00:19:27.040 align:middle line:84% So you've seen version of this before, 00:19:27.040 --> 00:19:30.270 align:middle line:84% and I remember asking you about it in the follow-up. 00:19:30.270 --> 00:19:33.450 align:middle line:90% We've got revenues and costs. 00:19:33.450 --> 00:19:37.770 align:middle line:84% In this case, this guy does a little bit of livestock, 00:19:37.770 --> 00:19:40.470 align:middle line:90% but mostly, it's wage labor. 00:19:40.470 --> 00:19:44.610 align:middle line:84% And in principle, could have had interest income, but didn't. 00:19:44.610 --> 00:19:47.460 align:middle line:84% And on the cost side, most of these costs 00:19:47.460 --> 00:19:51.180 align:middle line:84% are associated with livestock and in debt 00:19:51.180 --> 00:19:54.400 align:middle line:84% and actually has an interest expense, 00:19:54.400 --> 00:19:58.140 align:middle line:84% and then does consumption and subtracts that off of income 00:19:58.140 --> 00:19:59.260 align:middle line:90% and has some savings. 00:19:59.260 --> 00:20:03.270 align:middle line:84% So that's the income statement of household A, which 00:20:03.270 --> 00:20:05.130 align:middle line:84% we have these kinds of statements 00:20:05.130 --> 00:20:07.890 align:middle line:90% for roughly 800 households. 00:20:07.890 --> 00:20:09.690 align:middle line:90% Here's what happens over time. 00:20:09.690 --> 00:20:12.810 align:middle line:84% That labor income, which is the dominant source of income, 00:20:12.810 --> 00:20:17.670 align:middle line:84% falls and then gradually increases back up again. 00:20:17.670 --> 00:20:20.010 align:middle line:84% Other components are largely zero, 00:20:20.010 --> 00:20:22.830 align:middle line:84% although they started the business 00:20:22.830 --> 00:20:24.600 align:middle line:90% in the last year or two. 00:20:24.600 --> 00:20:27.630 align:middle line:84% Cultivation rises and then falls again. 00:20:27.630 --> 00:20:30.450 align:middle line:84% So trying this, that, or the other, but largely 00:20:30.450 --> 00:20:32.220 align:middle line:90% labor income. 00:20:32.220 --> 00:20:35.110 align:middle line:84% This same household has a balance sheet, 00:20:35.110 --> 00:20:38.250 align:middle line:84% which has liabilities on the right-hand side 00:20:38.250 --> 00:20:40.470 align:middle line:84% and assets on the left-hand side. 00:20:40.470 --> 00:20:42.480 align:middle line:84% Since those things don't balance, 00:20:42.480 --> 00:20:45.210 align:middle line:84% hopefully liabilities are smaller than assets, 00:20:45.210 --> 00:20:47.260 align:middle line:90% and the difference is net worth. 00:20:47.260 --> 00:20:51.000 align:middle line:84% So we have liabilities and net worth on the right-hand side 00:20:51.000 --> 00:20:52.650 align:middle line:90% and assets on the left. 00:20:52.650 --> 00:20:56.490 align:middle line:84% Assets consist of financial assets like cash 00:20:56.490 --> 00:21:00.600 align:middle line:84% and could have been a little bit of deposits in banks. 00:21:00.600 --> 00:21:03.420 align:middle line:84% And those are financial assets, and then 00:21:03.420 --> 00:21:07.810 align:middle line:84% we have real assets like livestock and so on, 00:21:07.810 --> 00:21:10.590 align:middle line:84% inventories, and we have other fixed assets, 00:21:10.590 --> 00:21:12.785 align:middle line:90% including household assets, OK? 00:21:12.785 --> 00:21:15.440 align:middle line:84% And you're wondering where I'm going with this. 00:21:15.440 --> 00:21:17.630 align:middle line:84% I'm deliberately highlighting things 00:21:17.630 --> 00:21:20.300 align:middle line:84% that are going to come up momentarily. 00:21:20.300 --> 00:21:24.860 align:middle line:84% And again, the balance sheet over time for this household, 00:21:24.860 --> 00:21:29.270 align:middle line:84% you can see wealth is going up, mostly 00:21:29.270 --> 00:21:31.550 align:middle line:90% because assets are going up. 00:21:31.550 --> 00:21:33.650 align:middle line:84% But toward the end, when the household started 00:21:33.650 --> 00:21:35.960 align:middle line:90% that business, they borrowed. 00:21:35.960 --> 00:21:39.940 align:middle line:84% Liabilities went up, and net worth actually went down. 00:21:39.940 --> 00:21:42.820 align:middle line:90% But anyway, this is over time. 00:21:42.820 --> 00:21:44.740 align:middle line:84% Finally, we have the statement of cash flow, 00:21:44.740 --> 00:21:47.940 align:middle line:84% which I highlighted when we talked about Bitcoin. 00:21:47.940 --> 00:21:50.730 align:middle line:84% We've got cash flow from production and cash flow 00:21:50.730 --> 00:21:54.330 align:middle line:84% from financing investment and consumption, 00:21:54.330 --> 00:21:57.270 align:middle line:84% and that produces a measure of cash flow 00:21:57.270 --> 00:21:59.730 align:middle line:84% from the flows, which has to sync up 00:21:59.730 --> 00:22:01.950 align:middle line:84% with the change in the cash and the balance 00:22:01.950 --> 00:22:05.650 align:middle line:84% sheet, which it does in these data. 00:22:05.650 --> 00:22:08.220 align:middle line:84% So next, I want to talk about diversity 00:22:08.220 --> 00:22:10.710 align:middle line:84% across the households, especially 00:22:10.710 --> 00:22:14.610 align:middle line:84% for the ones who are actually firms, who have 00:22:14.610 --> 00:22:16.690 align:middle line:90% decided to go into business. 00:22:16.690 --> 00:22:20.880 align:middle line:84% The types of activities, cultivation, farming business 00:22:20.880 --> 00:22:25.020 align:middle line:84% like retail outlets, trucking stuff around livestock, fish 00:22:25.020 --> 00:22:26.310 align:middle line:90% and shrimp. 00:22:26.310 --> 00:22:30.060 align:middle line:84% And this is like a Cobb-Douglas production function written out 00:22:30.060 --> 00:22:30.900 align:middle line:90% in logs. 00:22:30.900 --> 00:22:34.950 align:middle line:84% So delta k is the intensity of the capital good, 00:22:34.950 --> 00:22:39.270 align:middle line:84% and delta l is the intensity of the labor good. 00:22:39.270 --> 00:22:42.700 align:middle line:84% And what we want to see are those factor intensities. 00:22:42.700 --> 00:22:47.430 align:middle line:84% So in fact, if you look, cultivation is labor-intensive. 00:22:47.430 --> 00:22:49.650 align:middle line:84% It will fit the definition we reviewed 00:22:49.650 --> 00:22:51.540 align:middle line:90% at the beginning of class. 00:22:51.540 --> 00:22:54.330 align:middle line:84% Uses relatively more labor than capital 00:22:54.330 --> 00:22:58.800 align:middle line:84% because of the coefficients delta l and delta k. 00:22:58.800 --> 00:23:00.780 align:middle line:84% On the other extreme, the fish and shrimp 00:23:00.780 --> 00:23:05.290 align:middle line:84% uses almost no labor at all and uses tons of capital. 00:23:05.290 --> 00:23:09.390 align:middle line:84% So we've got capital and labor-intensive activities. 00:23:09.390 --> 00:23:14.130 align:middle line:84% We can try to make it simpler and talk about basically 00:23:14.130 --> 00:23:18.220 align:middle line:84% agricultural goods versus others called manufacturing. 00:23:18.220 --> 00:23:21.315 align:middle line:84% It really means non-agricultural. 00:23:21.315 --> 00:23:25.470 align:middle line:84% OK, the other thing sitting up here is this epsilon. 00:23:25.470 --> 00:23:28.020 align:middle line:84% This is quote, "an error term," but it actually 00:23:28.020 --> 00:23:33.430 align:middle line:84% contains not just noisy data, but a measurement of talent. 00:23:33.430 --> 00:23:37.515 align:middle line:84% So if we took the average for a particular household i, 00:23:37.515 --> 00:23:40.030 align:middle line:84% we took the average of these epsilons, 00:23:40.030 --> 00:23:43.270 align:middle line:84% they don't have to average out to 0. 00:23:43.270 --> 00:23:47.260 align:middle line:84% They average out to something that is common to the sector 00:23:47.260 --> 00:23:50.260 align:middle line:84% that household i has engaged in, a bar, 00:23:50.260 --> 00:23:52.750 align:middle line:84% and some other household-specific talent 00:23:52.750 --> 00:23:54.100 align:middle line:90% called z, OK? 00:23:54.100 --> 00:23:57.550 align:middle line:84% So there's a sector and household-specific talent 00:23:57.550 --> 00:24:01.450 align:middle line:84% variable that's wrapped up into these epsilons. 00:24:01.450 --> 00:24:03.370 align:middle line:84% Obviously, if you're talented, the epsilons 00:24:03.370 --> 00:24:06.220 align:middle line:84% are going to tend to be high, meaning output 00:24:06.220 --> 00:24:08.680 align:middle line:84% is going to be high, even controlling 00:24:08.680 --> 00:24:10.370 align:middle line:90% for capital and labor. 00:24:10.370 --> 00:24:13.960 align:middle line:84% So if you looked across the sectors at, quote, 00:24:13.960 --> 00:24:18.020 align:middle line:84% "sectoral productivity," they are pretty similar. 00:24:18.020 --> 00:24:19.990 align:middle line:84% The highest one here is livestock, 00:24:19.990 --> 00:24:22.510 align:middle line:90% the lowest one is business. 00:24:22.510 --> 00:24:24.940 align:middle line:84% Now, in terms of the household-specific 00:24:24.940 --> 00:24:28.990 align:middle line:84% productivities, there's a whole cross sectional dispersion 00:24:28.990 --> 00:24:32.440 align:middle line:90% of these zi's in each sector. 00:24:32.440 --> 00:24:36.820 align:middle line:84% And so this sigma is a characterization of the-- 00:24:36.820 --> 00:24:39.760 align:middle line:84% a bit like a variance of the distribution. 00:24:39.760 --> 00:24:44.590 align:middle line:84% The point being that each household has a particular z 00:24:44.590 --> 00:24:47.200 align:middle line:84% depending on the sector in which they reside, 00:24:47.200 --> 00:24:50.470 align:middle line:84% and they could be very talented or not talented at all, 00:24:50.470 --> 00:24:53.270 align:middle line:84% even though they're engaged in that activity. 00:24:53.270 --> 00:24:55.120 align:middle line:84% And these are estimated in the data. 00:24:55.120 --> 00:24:57.220 align:middle line:84% You're looking at the actual numbers. 00:24:57.220 --> 00:25:00.610 align:middle line:84% The other aspect of heterogeneity here, in addition 00:25:00.610 --> 00:25:06.070 align:middle line:84% to household talent, would be their wealth or, specifically, 00:25:06.070 --> 00:25:07.820 align:middle line:90% their fixed assets. 00:25:07.820 --> 00:25:09.440 align:middle line:84% That's why I emphasized it when we 00:25:09.440 --> 00:25:10.850 align:middle line:90% went through the balance sheet. 00:25:10.850 --> 00:25:14.080 align:middle line:84% And you can see the cross-sectional distribution 00:25:14.080 --> 00:25:17.830 align:middle line:84% in Lopburi, which is near Bangkok, 00:25:17.830 --> 00:25:23.410 align:middle line:84% versus the cross sectional distribution in Buriram, which 00:25:23.410 --> 00:25:28.750 align:middle line:84% has a lot more mass over here to the left, meaning low assets. 00:25:28.750 --> 00:25:31.870 align:middle line:84% So indeed, in the data, Lopburi near Bangkok 00:25:31.870 --> 00:25:36.100 align:middle line:84% has higher average fixed assets than does Buriram out 00:25:36.100 --> 00:25:37.420 align:middle line:90% in the Northeast. 00:25:37.420 --> 00:25:40.130 align:middle line:84% But again, it is the whole distribution. 00:25:40.130 --> 00:25:43.900 align:middle line:84% So a given household could have relatively little wealth 00:25:43.900 --> 00:25:46.600 align:middle line:84% or have a lot of wealth, even in Buriram. 00:25:46.600 --> 00:25:50.840 align:middle line:84% This actually is consistent with the Pauson, Townsend, 00:25:50.840 --> 00:25:54.970 align:middle line:84% Karaivanov paper, which was an application of mechanism 00:25:54.970 --> 00:26:00.100 align:middle line:84% design, where we talked about the talent of being a firm. 00:26:00.100 --> 00:26:02.650 align:middle line:90% Theta, I think, we had it there. 00:26:02.650 --> 00:26:06.880 align:middle line:84% And also, we talked about assets separately. 00:26:06.880 --> 00:26:10.100 align:middle line:84% And indeed, we talked about a limited liability constraint, 00:26:10.100 --> 00:26:11.470 align:middle line:90% which is coming up. 00:26:11.470 --> 00:26:14.590 align:middle line:84% So we have this cash thing, which, 00:26:14.590 --> 00:26:17.470 align:middle line:84% as I was saying, from your previous lecture, 00:26:17.470 --> 00:26:20.200 align:middle line:84% you know cash holdings don't behave very well. 00:26:20.200 --> 00:26:22.900 align:middle line:84% So we just impose what we see in the data, 00:26:22.900 --> 00:26:26.170 align:middle line:84% that cash holdings, that part of wealth that they're 00:26:26.170 --> 00:26:28.930 align:middle line:90% putting into cash, is fixed. 00:26:28.930 --> 00:26:31.930 align:middle line:84% It turns out that consumption is related to that. 00:26:31.930 --> 00:26:35.950 align:middle line:84% We imposed that consumption is equal to an intercept 00:26:35.950 --> 00:26:38.740 align:middle line:84% plus profits less that intercept. 00:26:38.740 --> 00:26:42.170 align:middle line:84% And gamma here has to do with omega, 00:26:42.170 --> 00:26:45.220 align:middle line:84% which is the fraction of savings that put into cash. 00:26:45.220 --> 00:26:47.650 align:middle line:84% I didn't put the algebra in, which I probably 00:26:47.650 --> 00:26:50.410 align:middle line:84% should have done, except all I want to say from this 00:26:50.410 --> 00:26:54.760 align:middle line:84% is the cash part of savings is just fitted to the model. 00:26:54.760 --> 00:26:57.040 align:middle line:90% It's not something we derived. 00:26:57.040 --> 00:27:00.340 align:middle line:84% So now I want to go back to those accounts 00:27:00.340 --> 00:27:04.330 align:middle line:84% and show you where gross national product comes 00:27:04.330 --> 00:27:07.090 align:middle line:84% from, except that, in this case, it's 00:27:07.090 --> 00:27:10.990 align:middle line:84% going to be village-level product and income. 00:27:10.990 --> 00:27:15.700 align:middle line:84% And then we'll go on to flow of funds, the balance sheet, 00:27:15.700 --> 00:27:17.750 align:middle line:90% balance of payments, and so on. 00:27:17.750 --> 00:27:20.770 align:middle line:90% So the top you've seen before. 00:27:20.770 --> 00:27:24.910 align:middle line:84% That's the statement of income, revenues less costs, 00:27:24.910 --> 00:27:28.850 align:middle line:84% but now we start subtracting things from both sides. 00:27:28.850 --> 00:27:32.410 align:middle line:84% For example, to get rid of capital gains, capital losses, 00:27:32.410 --> 00:27:35.980 align:middle line:84% and indemnity, we subtract them off, 00:27:35.980 --> 00:27:38.230 align:middle line:84% and then they appear with the opposite sign 00:27:38.230 --> 00:27:40.480 align:middle line:90% on the left-hand side. 00:27:40.480 --> 00:27:43.690 align:middle line:84% Likewise, we have expenses from production, which 00:27:43.690 --> 00:27:45.520 align:middle line:90% we subtract off of both sides. 00:27:45.520 --> 00:27:48.820 align:middle line:84% Hence, it disappears on the left hand side and reappears 00:27:48.820 --> 00:27:51.550 align:middle line:84% on the right, and it's subtracted. 00:27:51.550 --> 00:27:57.620 align:middle line:84% So we end up with revenues less expenses in production, 00:27:57.620 --> 00:27:59.350 align:middle line:90% which we call output. 00:27:59.350 --> 00:28:03.140 align:middle line:84% And on the left-hand side, we have the disposition of output. 00:28:03.140 --> 00:28:05.290 align:middle line:90% Output is used to pay interest. 00:28:05.290 --> 00:28:09.070 align:middle line:84% It's used to pay interest for capital. 00:28:09.070 --> 00:28:13.150 align:middle line:84% Normally, you would see wages paid to labor over here, 00:28:13.150 --> 00:28:15.470 align:middle line:84% and I'll come back to that momentarily. 00:28:15.470 --> 00:28:18.910 align:middle line:90% And the residual is profits. 00:28:18.910 --> 00:28:25.360 align:middle line:84% So all output gets paid, in the sense of paying hired factors 00:28:25.360 --> 00:28:29.300 align:middle line:84% or what's left over accruing to the business owner. 00:28:29.300 --> 00:28:32.650 align:middle line:84% So this is a classic decomposition 00:28:32.650 --> 00:28:38.720 align:middle line:84% of the national income accounts, from production or from inputs. 00:28:38.720 --> 00:28:41.070 align:middle line:84% We're really interested in international trade, 00:28:41.070 --> 00:28:42.860 align:middle line:84% so we're really interested, here, 00:28:42.860 --> 00:28:47.120 align:middle line:84% in how much is labor getting paid, how much of the interest 00:28:47.120 --> 00:28:51.140 align:middle line:84% being paid to capital, and so on as a function of output, 00:28:51.140 --> 00:28:54.590 align:middle line:84% just a bit like the 2x2x2 lecture. 00:28:54.590 --> 00:28:56.660 align:middle line:84% So if we're looking at output, remember, 00:28:56.660 --> 00:29:00.110 align:middle line:84% there are four villages in each province that were 00:29:00.110 --> 00:29:02.360 align:middle line:90% selected for the monthly data. 00:29:02.360 --> 00:29:04.460 align:middle line:84% Here, you can actually see output 00:29:04.460 --> 00:29:06.980 align:middle line:90% going down in Chachoengsao-- 00:29:06.980 --> 00:29:08.990 align:middle line:90% wiggling, but going down-- 00:29:08.990 --> 00:29:12.950 align:middle line:84% whereas, in Buriram, output is going up. 00:29:12.950 --> 00:29:15.260 align:middle line:84% And these are the other two provinces. 00:29:15.260 --> 00:29:19.340 align:middle line:84% So another point to make is that output 00:29:19.340 --> 00:29:21.110 align:middle line:84% is different across different villages, 00:29:21.110 --> 00:29:23.390 align:middle line:84% even though they're nearby to one another. 00:29:23.390 --> 00:29:25.940 align:middle line:84% And also, you'll miss it if I don't point it out, 00:29:25.940 --> 00:29:29.540 align:middle line:84% probably, which is the scale in the Northeast, 00:29:29.540 --> 00:29:33.590 align:middle line:84% the maximum value is 6 and 5, whereas, in the central, 00:29:33.590 --> 00:29:36.170 align:middle line:90% the maximum is 15 and 12. 00:29:36.170 --> 00:29:40.160 align:middle line:84% So yeah, villages near Bangkok have higher output 00:29:40.160 --> 00:29:44.180 align:middle line:84% than villages in the Northeast, but there remains heterogeneity 00:29:44.180 --> 00:29:48.050 align:middle line:84% across the villages, and you see the ups and downs. 00:29:48.050 --> 00:29:51.050 align:middle line:84% The shrimp ponds got hit with a bad shock 00:29:51.050 --> 00:29:53.390 align:middle line:84% that we've talked about previously in the network 00:29:53.390 --> 00:29:55.730 align:middle line:90% lecture under risk-sharing. 00:29:55.730 --> 00:29:59.990 align:middle line:84% So that's part of the decline in output in Chachoengsao. 00:29:59.990 --> 00:30:03.050 align:middle line:84% Buriram used to be a bunch of villages. 00:30:03.050 --> 00:30:07.740 align:middle line:84% We put rain gauges out there to measure waterfall and so on. 00:30:07.740 --> 00:30:11.510 align:middle line:84% And then it experienced a boom, and those villages 00:30:11.510 --> 00:30:13.310 align:middle line:90% were subsumed by the town. 00:30:13.310 --> 00:30:16.160 align:middle line:84% And you can see that expansion pretty dramatically 00:30:16.160 --> 00:30:18.050 align:middle line:90% in this output diagram. 00:30:18.050 --> 00:30:21.770 align:middle line:84% And as I was saying, we can decompose output 00:30:21.770 --> 00:30:23.420 align:middle line:90% into who gets it. 00:30:23.420 --> 00:30:28.760 align:middle line:84% And you can see here, wages, which is the darkest line, 00:30:28.760 --> 00:30:31.670 align:middle line:84% is the bigger part of the disposition 00:30:31.670 --> 00:30:33.050 align:middle line:90% of output in Buriram. 00:30:33.050 --> 00:30:36.740 align:middle line:84% In Sisaket, wages are not as much-- 00:30:36.740 --> 00:30:41.390 align:middle line:84% they're not trivial near Bangkok-- 00:30:41.390 --> 00:30:42.950 align:middle line:90% Chachoengsao and Lopburi. 00:30:42.950 --> 00:30:46.670 align:middle line:84% Instead, we have these two sectors, 00:30:46.670 --> 00:30:49.760 align:middle line:84% the labor-intensive agricultural sector 00:30:49.760 --> 00:30:53.930 align:middle line:84% and the capital-intensive manufacturing sector 00:30:53.930 --> 00:30:56.750 align:middle line:84% as sometimes having profit shares which 00:30:56.750 --> 00:30:59.030 align:middle line:90% are higher than the wages. 00:30:59.030 --> 00:31:01.940 align:middle line:84% Now we come to the savings investment account. 00:31:01.940 --> 00:31:04.590 align:middle line:84% We take the balance sheet at a point in time, 00:31:04.590 --> 00:31:06.360 align:middle line:84% and we take the first difference. 00:31:06.360 --> 00:31:09.230 align:middle line:84% So now we'll have the change in liabilities, the change the net 00:31:09.230 --> 00:31:12.650 align:middle line:84% worth on the right, change in financial and real physical 00:31:12.650 --> 00:31:14.520 align:middle line:90% assets on the left. 00:31:14.520 --> 00:31:18.020 align:middle line:84% We now go through this thing about subtracting items 00:31:18.020 --> 00:31:20.220 align:middle line:84% on the left and the right-hand side. 00:31:20.220 --> 00:31:23.540 align:middle line:84% We leave the change in net worth on the right-hand side 00:31:23.540 --> 00:31:25.670 align:middle line:84% and the disposition of that change 00:31:25.670 --> 00:31:30.140 align:middle line:84% into an increase in financial or real assets 00:31:30.140 --> 00:31:33.540 align:middle line:90% or a decrease in liabilities. 00:31:33.540 --> 00:31:37.040 align:middle line:84% So we can see they're managing their portfolio, 00:31:37.040 --> 00:31:39.732 align:middle line:84% in an accounting sense, as a function of, 00:31:39.732 --> 00:31:40.940 align:middle line:90% what do they do with savings. 00:31:40.940 --> 00:31:44.990 align:middle line:84% And likewise, if savings are negative because they consume 00:31:44.990 --> 00:31:48.710 align:middle line:84% more than they earn, then they have to draw that value out 00:31:48.710 --> 00:31:51.120 align:middle line:90% of something or borrow. 00:31:51.120 --> 00:31:53.480 align:middle line:84% So this is the allocation of savings 00:31:53.480 --> 00:31:55.430 align:middle line:90% across the four provinces. 00:31:55.430 --> 00:31:59.480 align:middle line:84% The darkest line here is savings, which is largely going 00:31:59.480 --> 00:32:01.370 align:middle line:90% up, although not entirely. 00:32:01.370 --> 00:32:03.800 align:middle line:84% Savings is going down and Chachoengsao, 00:32:03.800 --> 00:32:05.600 align:middle line:84% as you might have expected, since they 00:32:05.600 --> 00:32:08.810 align:middle line:84% got to finance those losses, that declining 00:32:08.810 --> 00:32:10.040 align:middle line:90% output from somewhere. 00:32:10.040 --> 00:32:13.700 align:middle line:84% They're preserving consumption by taking it out of savings. 00:32:13.700 --> 00:32:18.230 align:middle line:84% Likewise, they can save in the fixed assets, 00:32:18.230 --> 00:32:20.390 align:middle line:90% save in the financial assets. 00:32:20.390 --> 00:32:23.900 align:middle line:84% The thing that tracks the best overall-- 00:32:23.900 --> 00:32:27.020 align:middle line:84% tracks dark savings overall, the best-- 00:32:27.020 --> 00:32:28.070 align:middle line:90% is financial. 00:32:28.070 --> 00:32:31.220 align:middle line:84% And you can see in most of these diagrams, 00:32:31.220 --> 00:32:35.520 align:middle line:84% the black dotted lines are close to the solid black line. 00:32:35.520 --> 00:32:40.430 align:middle line:84% So financial assets increase when savings is positive, 00:32:40.430 --> 00:32:43.670 align:middle line:84% but these other real and physical stuff 00:32:43.670 --> 00:32:45.500 align:middle line:90% as well as gifts play a role. 00:32:45.500 --> 00:32:48.290 align:middle line:84% Gifts are not trivial in these economies, 00:32:48.290 --> 00:32:51.880 align:middle line:84% as you already know from thinking about insurance and so 00:32:51.880 --> 00:32:52.380 align:middle line:90% on. 00:32:52.380 --> 00:32:55.610 align:middle line:90% So all these things add up. 00:32:55.610 --> 00:32:58.250 align:middle line:84% Gifts, financial transfers, inventory, 00:32:58.250 --> 00:33:00.410 align:middle line:90% and so on add up to savings. 00:33:00.410 --> 00:33:04.290 align:middle line:84% But gifts could be negative, because they're not incoming, 00:33:04.290 --> 00:33:05.600 align:middle line:90% but they're outgoing. 00:33:05.600 --> 00:33:06.620 align:middle line:90% They're not saving it. 00:33:06.620 --> 00:33:08.090 align:middle line:90% They're giving it away. 00:33:08.090 --> 00:33:09.560 align:middle line:90% It's not always negative. 00:33:09.560 --> 00:33:12.000 align:middle line:84% Gifts are positive, in some instances. 00:33:12.000 --> 00:33:16.740 align:middle line:84% But in any event, they're not a trivial order of magnitude, 00:33:16.740 --> 00:33:19.430 align:middle line:84% and we will look at that subsequently again. 00:33:19.430 --> 00:33:22.520 align:middle line:84% In terms of financial flows, you've 00:33:22.520 --> 00:33:25.670 align:middle line:84% got something called the current account. 00:33:25.670 --> 00:33:29.060 align:middle line:84% For the US, for example, are they running a trade balance, 00:33:29.060 --> 00:33:31.460 align:middle line:84% or are they earning money from abroad? 00:33:31.460 --> 00:33:33.860 align:middle line:84% And likewise, for these villages, 00:33:33.860 --> 00:33:36.110 align:middle line:90% they export and import stuff. 00:33:36.110 --> 00:33:39.960 align:middle line:84% We take the trade balance as the difference if positive. 00:33:39.960 --> 00:33:41.610 align:middle line:90% Exports exceed imports. 00:33:41.610 --> 00:33:44.430 align:middle line:84% And likewise, some member of the household 00:33:44.430 --> 00:33:47.670 align:middle line:84% is in Bangkok and earning money and sending that money 00:33:47.670 --> 00:33:52.800 align:middle line:84% back home as a gift, that's factor income earned abroad. 00:33:52.800 --> 00:33:55.740 align:middle line:84% In China, people work in the Philippines 00:33:55.740 --> 00:33:57.360 align:middle line:90% and send money back home. 00:33:57.360 --> 00:33:59.950 align:middle line:84% So this is called the current account. 00:33:59.950 --> 00:34:04.320 align:middle line:84% Now, there are corresponding items, offsetting items, 00:34:04.320 --> 00:34:06.630 align:middle line:90% on the financial account. 00:34:06.630 --> 00:34:09.630 align:middle line:84% Basically, if you're running a trade surplus, 00:34:09.630 --> 00:34:12.210 align:middle line:84% you're acquiring claims on foreigners, 00:34:12.210 --> 00:34:14.790 align:middle line:84% which could be by holding their currency, 00:34:14.790 --> 00:34:18.210 align:middle line:84% holding their financial assets, or holding their real assets. 00:34:18.210 --> 00:34:20.880 align:middle line:84% And as we'll talk about next time, 00:34:20.880 --> 00:34:23.370 align:middle line:84% the US has been running a trade deficit 00:34:23.370 --> 00:34:27.989 align:middle line:84% with China, which means China is acquiring claims on the US, 00:34:27.989 --> 00:34:31.929 align:middle line:84% holding US treasuries, real estate, and so on. 00:34:31.929 --> 00:34:34.350 align:middle line:84% So the names change to protect the innocent, 00:34:34.350 --> 00:34:36.900 align:middle line:84% in the sense of we're not looking at countries here. 00:34:36.900 --> 00:34:39.690 align:middle line:84% We're looking at villages, but we're doing all the accounting 00:34:39.690 --> 00:34:42.090 align:middle line:84% for the villages the way it's done 00:34:42.090 --> 00:34:46.409 align:middle line:84% for countries, with exactly the same accounting identities. 00:34:46.409 --> 00:34:49.469 align:middle line:84% And the point of all of this is only to tell you 00:34:49.469 --> 00:34:51.570 align:middle line:84% where some of these things that you've heard of 00:34:51.570 --> 00:34:52.530 align:middle line:90% are coming from. 00:34:52.530 --> 00:34:55.590 align:middle line:84% From the ground up, from each and every household, 00:34:55.590 --> 00:34:58.680 align:middle line:84% we have the financial accounts-- income statement, 00:34:58.680 --> 00:35:00.180 align:middle line:90% balance sheet, and so on. 00:35:00.180 --> 00:35:03.960 align:middle line:84% We can add it up over all households in the village, 00:35:03.960 --> 00:35:09.550 align:middle line:84% as the sample allows, and get the village-level aggregates. 00:35:09.550 --> 00:35:11.520 align:middle line:90% This is for all of Thailand. 00:35:11.520 --> 00:35:15.420 align:middle line:84% Thailand was running a trade deficit, 00:35:15.420 --> 00:35:18.780 align:middle line:84% and foreigners were acquiring claims on Thailand 00:35:18.780 --> 00:35:23.550 align:middle line:84% until the 1997 financial crisis, in which case, things reversed. 00:35:23.550 --> 00:35:26.940 align:middle line:84% And they've been a bit unstable ever since. 00:35:26.940 --> 00:35:29.310 align:middle line:84% In this case, balance of payments 00:35:29.310 --> 00:35:33.630 align:middle line:84% are roughly a certain of GDP, maybe 10%. 00:35:33.630 --> 00:35:34.530 align:middle line:90% That's a big number. 00:35:34.530 --> 00:35:36.990 align:middle line:90% It's typically 5%. 00:35:36.990 --> 00:35:39.720 align:middle line:84% So when we come to villages, you might ask, 00:35:39.720 --> 00:35:41.740 align:middle line:90% how big are these flows? 00:35:41.740 --> 00:35:44.490 align:middle line:84% And the answer is, typically, much bigger. 00:35:44.490 --> 00:35:48.450 align:middle line:84% For example, in Chachoengsao, the balance of payments, 00:35:48.450 --> 00:35:51.060 align:middle line:84% including the trade and the current account, 00:35:51.060 --> 00:35:54.510 align:middle line:90% can be as high as 20% of GDP. 00:35:54.510 --> 00:35:58.450 align:middle line:84% And Lopburi, likewise reaches 15% of GDP. 00:35:58.450 --> 00:36:00.820 align:middle line:84% So they are small open economies. 00:36:00.820 --> 00:36:05.400 align:middle line:84% These villages are engaged with the rest of the country. 00:36:05.400 --> 00:36:07.260 align:middle line:84% And in fact, we can break that down 00:36:07.260 --> 00:36:10.380 align:middle line:84% into the gifts, which I showed you a moment ago, 00:36:10.380 --> 00:36:15.030 align:middle line:84% as well as, here, the measure of openness 00:36:15.030 --> 00:36:18.850 align:middle line:84% in terms of trade, which is exports minus imports. 00:36:18.850 --> 00:36:22.860 align:middle line:84% So roughly, Chachoengsao near Bangkok, its imports 00:36:22.860 --> 00:36:27.370 align:middle line:84% and exports are almost matching one another. 00:36:27.370 --> 00:36:30.420 align:middle line:84% Whereas, out in the Northeast, the imports 00:36:30.420 --> 00:36:33.450 align:middle line:84% are a lot bigger than the exports. 00:36:33.450 --> 00:36:35.610 align:middle line:90% The imports are for production. 00:36:35.610 --> 00:36:38.160 align:middle line:84% They're inputs acquired for production, 00:36:38.160 --> 00:36:41.160 align:middle line:84% and they export a little bit of their output. 00:36:41.160 --> 00:36:46.330 align:middle line:84% They eat a lot of the rice that they produce and export less. 00:36:46.330 --> 00:36:49.210 align:middle line:84% I'm getting ready to show you the model. 00:36:49.210 --> 00:36:51.760 align:middle line:84% One, what about the relative price? 00:36:51.760 --> 00:36:53.290 align:middle line:84% At the beginning of this class, we 00:36:53.290 --> 00:36:57.070 align:middle line:84% talked about the price of good 1 the price of good 2. 00:36:57.070 --> 00:37:00.370 align:middle line:84% We have two sectors for simplicity, agriculture 00:37:00.370 --> 00:37:01.960 align:middle line:90% and manufacturing. 00:37:01.960 --> 00:37:05.380 align:middle line:84% So from the data, we can actually see, overall, 00:37:05.380 --> 00:37:08.590 align:middle line:84% that the price of the agricultural good is going up, 00:37:08.590 --> 00:37:12.130 align:middle line:84% the ratio is going down, from the standpoint 00:37:12.130 --> 00:37:16.180 align:middle line:84% of a typical village, with the possibility of imports 00:37:16.180 --> 00:37:20.800 align:middle line:84% and exports, just the way we had countries possibly exporting 00:37:20.800 --> 00:37:22.720 align:middle line:90% and importing to each other. 00:37:22.720 --> 00:37:26.890 align:middle line:84% In terms of the factor prices, remember the minimized cost 00:37:26.890 --> 00:37:27.520 align:middle line:90% curve? 00:37:27.520 --> 00:37:29.150 align:middle line:90% Well, this is over time. 00:37:29.150 --> 00:37:32.410 align:middle line:84% So we can see the wage rate and the interest rate, 00:37:32.410 --> 00:37:36.010 align:middle line:84% both adjusted to be in units of consumption. 00:37:36.010 --> 00:37:38.800 align:middle line:84% Hence, real values as a numerator. 00:37:38.800 --> 00:37:44.920 align:middle line:84% So in the areas near Bangkok, the blue and the red, 00:37:44.920 --> 00:37:48.940 align:middle line:84% we can see that wages tend to be higher 00:37:48.940 --> 00:37:51.430 align:middle line:90% than they are in the Northeast. 00:37:51.430 --> 00:37:53.680 align:middle line:84% Now, near Bangkok, you would think 00:37:53.680 --> 00:37:56.050 align:middle line:84% this is relatively capital-abundant, 00:37:56.050 --> 00:37:59.440 align:middle line:84% because they're pretty industrialized and short 00:37:59.440 --> 00:38:00.850 align:middle line:90% of labor. 00:38:00.850 --> 00:38:03.970 align:middle line:84% So indeed, the relative price of labor to capital 00:38:03.970 --> 00:38:08.260 align:middle line:84% is higher in the areas near Bangkok, 00:38:08.260 --> 00:38:11.350 align:middle line:84% reflecting the capital abundance and the relative shortage 00:38:11.350 --> 00:38:12.310 align:middle line:90% of labor. 00:38:12.310 --> 00:38:14.770 align:middle line:84% The reverse is true in the Northeast. 00:38:14.770 --> 00:38:17.530 align:middle line:84% And also, we can look at the interest rate, which 00:38:17.530 --> 00:38:19.450 align:middle line:90% is like the price of capital. 00:38:19.450 --> 00:38:24.280 align:middle line:84% And again, you can see that now the blue and the red 00:38:24.280 --> 00:38:27.490 align:middle line:84% are at the bottom, because places near Bangkok 00:38:27.490 --> 00:38:29.080 align:middle line:90% are capital-abundant. 00:38:29.080 --> 00:38:32.090 align:middle line:84% And so the interest rates tend to be low. 00:38:32.090 --> 00:38:35.110 align:middle line:84% We did this experiment for the Heckscher-Ohlin model. 00:38:35.110 --> 00:38:37.780 align:middle line:84% Countries that are relatively abundant in a factor 00:38:37.780 --> 00:38:41.410 align:middle line:84% should be exporting the good which uses that factor 00:38:41.410 --> 00:38:43.340 align:middle line:90% relatively intensively. 00:38:43.340 --> 00:38:47.500 align:middle line:84% So you can imagine that places near Bangkok 00:38:47.500 --> 00:38:50.080 align:middle line:84% that are abundant in capital will have a tendency 00:38:50.080 --> 00:38:52.390 align:middle line:84% to export the manufactured goods, 00:38:52.390 --> 00:38:55.210 align:middle line:84% because that's the one that's intensive in capital. 00:38:55.210 --> 00:38:57.100 align:middle line:90% What am I doing today? 00:38:57.100 --> 00:39:01.990 align:middle line:84% I'm just getting the labels to be realistic in actual context, 00:39:01.990 --> 00:39:04.645 align:middle line:90% realistic relative to last time. 00:39:04.645 --> 00:39:06.040 align:middle line:90% We learned the theory first. 00:39:06.040 --> 00:39:09.340 align:middle line:84% Now we're applying it, although with some modifications. 00:39:09.340 --> 00:39:11.860 align:middle line:84% The other thing you can see here is the interest 00:39:11.860 --> 00:39:13.420 align:middle line:90% rates are converging. 00:39:13.420 --> 00:39:16.310 align:middle line:84% They're getting really close to one another over time. 00:39:16.310 --> 00:39:20.260 align:middle line:84% So one imagines that if they could [INAUDIBLE] 00:39:20.260 --> 00:39:21.790 align:middle line:90% trade in inputs. 00:39:21.790 --> 00:39:23.140 align:middle line:90% Not just outputs. 00:39:23.140 --> 00:39:25.840 align:middle line:84% It's pretty easy to do that with a capital input. 00:39:25.840 --> 00:39:28.730 align:middle line:84% But the wage rates are not converging, 00:39:28.730 --> 00:39:31.810 align:middle line:84% so that's also a hint that, like the 2x2 model, 00:39:31.810 --> 00:39:37.260 align:middle line:84% we're not going to allow labor to be moving across borders. 00:39:37.260 --> 00:39:41.387 align:middle line:84% OK, so I'm going to show you the ingredients for a small open, 00:39:41.387 --> 00:39:42.970 align:middle line:84% economy and then I'm going to tell you 00:39:42.970 --> 00:39:46.910 align:middle line:84% what the difference is with a 2x2x2 that we did last time. 00:39:46.910 --> 00:39:49.180 align:middle line:90% So it is 2x2-- 00:39:49.180 --> 00:39:51.460 align:middle line:90% 2 goods, 2 factors-- 00:39:51.460 --> 00:39:54.958 align:middle line:84% although it's a village relative to the rest of the country. 00:39:54.958 --> 00:39:56.500 align:middle line:84% Different from last time, we're going 00:39:56.500 --> 00:40:00.640 align:middle line:84% to have financial frictions, like those borrowing-to-wealth 00:40:00.640 --> 00:40:01.900 align:middle line:90% ratios. 00:40:01.900 --> 00:40:05.050 align:middle line:84% There's a lot of agents in each village, 00:40:05.050 --> 00:40:08.200 align:middle line:84% and they differ in their wealth and ability levels. 00:40:08.200 --> 00:40:12.130 align:middle line:84% They can choose to be workers or they could become a firm. 00:40:12.130 --> 00:40:13.990 align:middle line:84% And if they're an entrepreneur, they 00:40:13.990 --> 00:40:17.860 align:middle line:84% could be in the manufacturing or agricultural sector. 00:40:17.860 --> 00:40:21.090 align:middle line:90% So that's one of the two 2s-- 00:40:21.090 --> 00:40:26.260 align:middle line:84% two goods, agriculture and manufacturing, two inputs, 00:40:26.260 --> 00:40:30.010 align:middle line:84% capital and labor, with the factor intensities different. 00:40:30.010 --> 00:40:35.260 align:middle line:84% Labor is owned by households and supplied inelastically, 00:40:35.260 --> 00:40:37.210 align:middle line:90% like it was last time. 00:40:37.210 --> 00:40:39.340 align:middle line:90% They get this wage w. 00:40:39.340 --> 00:40:44.770 align:middle line:84% Wealth is owned by households and supplied at an interest 00:40:44.770 --> 00:40:47.800 align:middle line:84% rate, but it's also changing over time. 00:40:47.800 --> 00:40:50.680 align:middle line:84% Unlike what we had last time, we've 00:40:50.680 --> 00:40:54.700 align:middle line:84% got dynamics in a way to save money over time. 00:40:54.700 --> 00:40:57.940 align:middle line:84% They can use their capital, or they can lend it out. 00:40:57.940 --> 00:41:00.310 align:middle line:84% So we have a borrowing and lending market, 00:41:00.310 --> 00:41:02.920 align:middle line:84% but that capital market is imperfect. 00:41:02.920 --> 00:41:07.420 align:middle line:84% Namely, you can't necessarily use an unlimited amount 00:41:07.420 --> 00:41:08.980 align:middle line:90% of capital k. 00:41:08.980 --> 00:41:11.660 align:middle line:90% You're bounded by your wealth. 00:41:11.660 --> 00:41:14.540 align:middle line:84% So that, for the third time today, 00:41:14.540 --> 00:41:17.970 align:middle line:84% is mention of the limited liability constraints. 00:41:17.970 --> 00:41:20.810 align:middle line:84% So the differences with last time, 00:41:20.810 --> 00:41:23.630 align:middle line:84% we have decreasing returns to scale, 00:41:23.630 --> 00:41:26.120 align:middle line:84% whereas, last time, in the theory lecture, 00:41:26.120 --> 00:41:29.180 align:middle line:84% we used and abused the fact that it 00:41:29.180 --> 00:41:31.610 align:middle line:90% was constant returns to scale-- 00:41:31.610 --> 00:41:34.730 align:middle line:84% hence, unit costs of production, which 00:41:34.730 --> 00:41:37.820 align:middle line:84% scale up and down depending on the level of production. 00:41:37.820 --> 00:41:41.870 align:middle line:84% Factor mobility is somewhat limited here, 00:41:41.870 --> 00:41:45.500 align:middle line:84% but the capital good moves around, unlike last time. 00:41:45.500 --> 00:41:49.850 align:middle line:84% We have financial variables here and real variables, 00:41:49.850 --> 00:41:53.300 align:middle line:84% whereas, last time, we only had real variables. 00:41:53.300 --> 00:41:56.420 align:middle line:84% And likewise, here, we have borrowing constraints. 00:41:56.420 --> 00:41:59.160 align:middle line:84% And last time, we didn't have any of that. 00:41:59.160 --> 00:42:02.790 align:middle line:84% But this model does have the different endowments 00:42:02.790 --> 00:42:04.120 align:middle line:90% of the factors. 00:42:04.120 --> 00:42:07.620 align:middle line:84% It does have the different factor intensities. 00:42:07.620 --> 00:42:10.020 align:middle line:90% It keeps the 2x2x2-- 00:42:10.020 --> 00:42:13.320 align:middle line:84% well, 2x2 anyway with small open economy. 00:42:13.320 --> 00:42:17.700 align:middle line:84% And so part of what you're going to see in the simulations 00:42:17.700 --> 00:42:21.160 align:middle line:84% reflects what we learned last time. 00:42:21.160 --> 00:42:23.010 align:middle line:84% And part of it is a bit different, 00:42:23.010 --> 00:42:29.070 align:middle line:84% because it has a few additional realistic features. 00:42:29.070 --> 00:42:32.250 align:middle line:84% Rather than take you through the entire model, 00:42:32.250 --> 00:42:36.000 align:middle line:84% I'll just give you the notation at the beginning. 00:42:36.000 --> 00:42:40.410 align:middle line:84% Each entrepreneur can produce sector a, 00:42:40.410 --> 00:42:43.680 align:middle line:84% which could be agriculture, but it's also manufacturing. 00:42:43.680 --> 00:42:48.690 align:middle line:84% A household has a certain talent level z and an initial capital 00:42:48.690 --> 00:42:49.470 align:middle line:90% stock. 00:42:49.470 --> 00:42:55.140 align:middle line:84% It will use its inputs, capital and higher labor, 00:42:55.140 --> 00:42:58.080 align:middle line:84% at its productivity level, selling it 00:42:58.080 --> 00:42:59.880 align:middle line:90% for the price of output. 00:42:59.880 --> 00:43:03.780 align:middle line:84% But it has to pay input costs for capital and labor, 00:43:03.780 --> 00:43:06.510 align:middle line:84% and the input prices are r and w. 00:43:06.510 --> 00:43:08.760 align:middle line:90% So you've seen this many times. 00:43:08.760 --> 00:43:11.850 align:middle line:84% This is just a statement of profits for a firm. 00:43:11.850 --> 00:43:15.090 align:middle line:84% It happens to be a firm owned and entirely operated 00:43:15.090 --> 00:43:19.560 align:middle line:84% by a given household, because the theta ij equals 1. 00:43:19.560 --> 00:43:21.840 align:middle line:84% And here's the borrowing constraint. 00:43:21.840 --> 00:43:26.280 align:middle line:84% Capital utilized is limited by the pre-existing stock 00:43:26.280 --> 00:43:29.160 align:middle line:84% of fixed assets [INAUDIBLE] this constant c 00:43:29.160 --> 00:43:31.380 align:middle line:84% could be greater than 1, which allows 00:43:31.380 --> 00:43:36.210 align:middle line:84% them to borrow, to acquire fixed assets and production greater 00:43:36.210 --> 00:43:38.640 align:middle line:90% than what they already own. 00:43:38.640 --> 00:43:40.950 align:middle line:90% Labor is non-negative. 00:43:40.950 --> 00:43:44.310 align:middle line:84% As in that paper with Anna Palson and Karaivanov 00:43:44.310 --> 00:43:47.820 align:middle line:84% they could choose to be a wage earner, put all their wealth-- 00:43:47.820 --> 00:43:51.390 align:middle line:84% lend it out, basically, earn interest on it, 00:43:51.390 --> 00:43:54.240 align:middle line:84% or they earn money from being a wage earner. 00:43:54.240 --> 00:43:56.850 align:middle line:84% Likewise, we have profits for agriculturalists 00:43:56.850 --> 00:43:58.470 align:middle line:90% and manufacturers. 00:43:58.470 --> 00:44:02.130 align:middle line:84% And they're, in principle, lending stuff out, 00:44:02.130 --> 00:44:04.320 align:middle line:84% but they could be borrowing some of it 00:44:04.320 --> 00:44:06.460 align:middle line:90% back to finance production. 00:44:06.460 --> 00:44:09.210 align:middle line:84% And then we have a utility over consumption 00:44:09.210 --> 00:44:12.360 align:middle line:84% of agriculture and manufactured goods, which 00:44:12.360 --> 00:44:17.310 align:middle line:84% looks this Cobb-Douglas thing, where these etas are cleverly 00:44:17.310 --> 00:44:21.270 align:middle line:84% chosen to allow us to talk about aggregate savings. 00:44:21.270 --> 00:44:22.920 align:middle line:84% Because, in fact, there's capital 00:44:22.920 --> 00:44:26.460 align:middle line:84% in manufacturing and capital in agriculture, 00:44:26.460 --> 00:44:28.260 align:middle line:84% and we don't want to keep having to refer 00:44:28.260 --> 00:44:31.080 align:middle line:84% to the distinction about where they're saving. 00:44:31.080 --> 00:44:33.690 align:middle line:84% So we come up with this ideal price index. 00:44:33.690 --> 00:44:35.570 align:middle line:90% Don't worry about it too much. 00:44:35.570 --> 00:44:36.600 align:middle line:90% It exists. 00:44:36.600 --> 00:44:39.934 align:middle line:90% That's all we need to know. 00:44:39.934 --> 00:44:43.270 align:middle line:90% OK, it's a small open economy. 00:44:43.270 --> 00:44:46.090 align:middle line:84% Households have their capital endowment. 00:44:46.090 --> 00:44:49.510 align:middle line:84% We fit a distribution that fits pretty well, 00:44:49.510 --> 00:44:52.330 align:middle line:84% relative to the histograms I've already shown you. 00:44:52.330 --> 00:44:54.700 align:middle line:90% We have the labor endowments. 00:44:54.700 --> 00:44:56.900 align:middle line:90% We know how much time they have. 00:44:56.900 --> 00:44:58.930 align:middle line:84% We give them a little bit of sleep 00:44:58.930 --> 00:45:01.090 align:middle line:84% and come up with how many units of labor 00:45:01.090 --> 00:45:05.290 align:middle line:84% they can supply over the year, which is fixed and exogenous. 00:45:05.290 --> 00:45:06.830 align:middle line:90% Then they have a savings rate. 00:45:06.830 --> 00:45:10.150 align:middle line:84% Which, again, as I already said, that was an omega. 00:45:10.150 --> 00:45:12.760 align:middle line:90% We fit to match the data. 00:45:12.760 --> 00:45:16.690 align:middle line:84% I mentioned that, in the context of the cash holdings, 00:45:16.690 --> 00:45:19.720 align:middle line:84% one could imagine having these households solve 00:45:19.720 --> 00:45:22.510 align:middle line:84% discounted expected utility maximization 00:45:22.510 --> 00:45:25.210 align:middle line:84% problems over their lifetime that 00:45:25.210 --> 00:45:27.320 align:middle line:90% adds a lot to the complication. 00:45:27.320 --> 00:45:30.550 align:middle line:84% So instead of doing that, we just basically impose 00:45:30.550 --> 00:45:33.550 align:middle line:84% from the data onto the households 00:45:33.550 --> 00:45:35.900 align:middle line:90% an estimated savings rate. 00:45:35.900 --> 00:45:38.050 align:middle line:84% Now, this is a kind of interesting diagram, 00:45:38.050 --> 00:45:42.460 align:middle line:84% because all these parameters, like the factor intensity 00:45:42.460 --> 00:45:45.490 align:middle line:84% numbers and everything else, are going 00:45:45.490 --> 00:45:49.780 align:middle line:84% to be calibrated, which I'll tell you about momentarily. 00:45:49.780 --> 00:45:53.750 align:middle line:84% And we will generate pictures like this. 00:45:53.750 --> 00:45:56.830 align:middle line:84% So I mentioned the calibration, only because you 00:45:56.830 --> 00:45:58.690 align:middle line:90% can draw any picture you want. 00:45:58.690 --> 00:46:01.690 align:middle line:84% But this is what it's going to look like after the work is 00:46:01.690 --> 00:46:03.640 align:middle line:90% done of estimation. 00:46:03.640 --> 00:46:06.550 align:middle line:84% Over here, we have low-wealth households 00:46:06.550 --> 00:46:08.830 align:middle line:90% versus high-wealth households. 00:46:08.830 --> 00:46:10.690 align:middle line:84% The other dimension of heterogeneity 00:46:10.690 --> 00:46:14.230 align:middle line:84% is low talent versus high talent. 00:46:14.230 --> 00:46:17.002 align:middle line:84% And you can see if they're not very talented, 00:46:17.002 --> 00:46:19.210 align:middle line:84% they're not going to go into either type of business. 00:46:19.210 --> 00:46:21.110 align:middle line:90% They're going to be workers. 00:46:21.110 --> 00:46:25.330 align:middle line:84% On the other hand, if talent reaches a higher level, 00:46:25.330 --> 00:46:29.650 align:middle line:84% they could decide to go into manufacturing or agriculture. 00:46:29.650 --> 00:46:32.500 align:middle line:84% And then for a given level of wealth, 00:46:32.500 --> 00:46:35.530 align:middle line:84% they may be constrained or unconstrained. 00:46:35.530 --> 00:46:38.440 align:middle line:84% Ironically, the more talented they are, 00:46:38.440 --> 00:46:41.502 align:middle line:84% given a certain borrowing constraint and their wealth, 00:46:41.502 --> 00:46:43.210 align:middle line:84% the more constrained they're going to be, 00:46:43.210 --> 00:46:44.627 align:middle line:84% because they would like to do more 00:46:44.627 --> 00:46:46.840 align:middle line:84% and they're not allowed to do it. 00:46:46.840 --> 00:46:49.660 align:middle line:84% And here, it actually moves from being 00:46:49.660 --> 00:46:52.600 align:middle line:84% constrained in manufacturing to being 00:46:52.600 --> 00:46:55.020 align:middle line:90% constrained in agriculture. 00:46:55.020 --> 00:46:57.240 align:middle line:84% So it's the really, really talented guys 00:46:57.240 --> 00:47:01.350 align:middle line:84% who are sticking on the farm, given the parameters that we 00:47:01.350 --> 00:47:02.920 align:middle line:90% estimated. 00:47:02.920 --> 00:47:04.100 align:middle line:90% This I've already said. 00:47:04.100 --> 00:47:06.670 align:middle line:84% We're going to have an open market for capital, 00:47:06.670 --> 00:47:09.310 align:middle line:90% but a closed market for labor. 00:47:09.310 --> 00:47:11.810 align:middle line:84% And again, that's because the interest rates are converging. 00:47:11.810 --> 00:47:13.570 align:middle line:84% So it looks like, in the end, at least, 00:47:13.570 --> 00:47:18.340 align:middle line:84% capital can be moved across provinces without a cost. 00:47:18.340 --> 00:47:21.400 align:middle line:84% Here's, finally, the calibration slide. 00:47:21.400 --> 00:47:24.970 align:middle line:84% We have to figure out the relative prices. 00:47:24.970 --> 00:47:26.860 align:middle line:90% We need the interest rate. 00:47:26.860 --> 00:47:29.080 align:middle line:84% We're going to take it as exogenous. 00:47:29.080 --> 00:47:31.420 align:middle line:84% The relative price of manufacturing 00:47:31.420 --> 00:47:34.810 align:middle line:84% versus agriculture we're going to take as exogenous 00:47:34.810 --> 00:47:39.940 align:middle line:84% and match it up with what we see in the data as exogenous. 00:47:39.940 --> 00:47:42.590 align:middle line:84% I'll come back to how we pick it in a minute. 00:47:42.590 --> 00:47:45.820 align:middle line:84% And the borrowing rates are exogenous. 00:47:45.820 --> 00:47:47.650 align:middle line:84% We stick in, for the interest rate, 00:47:47.650 --> 00:47:49.510 align:middle line:90% what we observe in the data. 00:47:49.510 --> 00:47:52.090 align:middle line:84% We don't know about this relative price of the borrowing 00:47:52.090 --> 00:47:57.670 align:middle line:84% limit, but we do have the profit sharing in each sector. 00:47:57.670 --> 00:48:02.180 align:middle line:84% So we move the relative prices to match the profit share. 00:48:02.180 --> 00:48:07.660 align:middle line:84% So for example, if agriculture has a price pa 00:48:07.660 --> 00:48:12.430 align:middle line:84% and we observe a large relative share of profits 00:48:12.430 --> 00:48:16.000 align:middle line:84% from agriculture, then we imagine 00:48:16.000 --> 00:48:19.870 align:middle line:84% that it's a relatively high price of the agricultural goods 00:48:19.870 --> 00:48:23.560 align:middle line:84% relative to the manufacturing goods and vice versa. 00:48:23.560 --> 00:48:27.280 align:middle line:84% Likewise, we're going to move the borrowing limit to match 00:48:27.280 --> 00:48:29.410 align:middle line:90% the observed wage rate. 00:48:29.410 --> 00:48:33.140 align:middle line:84% The point being that if they can't borrow very much, 00:48:33.140 --> 00:48:36.190 align:middle line:84% they're not going to have a very high scale of production. 00:48:36.190 --> 00:48:37.930 align:middle line:84% They're not going to hire a lot of labor. 00:48:37.930 --> 00:48:40.750 align:middle line:84% The demand for labor being low means 00:48:40.750 --> 00:48:43.840 align:middle line:84% the wage that has to clear the local market 00:48:43.840 --> 00:48:45.520 align:middle line:90% is going to be low. 00:48:45.520 --> 00:48:48.010 align:middle line:84% The bigger picture here is calibration. 00:48:48.010 --> 00:48:51.910 align:middle line:84% You have objects that matter a lot in the model. 00:48:51.910 --> 00:48:55.210 align:middle line:84% And how you pick them, well, here, we 00:48:55.210 --> 00:48:59.260 align:middle line:84% pick them by taking the data that we have 00:48:59.260 --> 00:49:02.110 align:middle line:84% and varying the parameters in the model in such a way 00:49:02.110 --> 00:49:05.020 align:middle line:84% that we can fit the model-generated data 00:49:05.020 --> 00:49:06.760 align:middle line:90% to the actual data. 00:49:06.760 --> 00:49:08.860 align:middle line:84% Doing that, we have to make a choice 00:49:08.860 --> 00:49:11.620 align:middle line:84% of which part of the actual data to use 00:49:11.620 --> 00:49:14.590 align:middle line:90% to calibrate which parameter. 00:49:14.590 --> 00:49:16.110 align:middle line:84% And again, read the Dawkins thing, 00:49:16.110 --> 00:49:20.210 align:middle line:84% and you'll see a discussion of this much more generally. 00:49:20.210 --> 00:49:24.180 align:middle line:84% OK, so let's look at what happens. 00:49:24.180 --> 00:49:25.940 align:middle line:90% How well are we doing? 00:49:25.940 --> 00:49:28.530 align:middle line:84% Oh, these economists, they come up with these models. 00:49:28.530 --> 00:49:31.710 align:middle line:84% God knows how well they predict, so I'll 00:49:31.710 --> 00:49:33.720 align:middle line:90% show you how well we're doing. 00:49:33.720 --> 00:49:38.340 align:middle line:84% This is, at the calibrated values for Lopburi, I believe-- 00:49:38.340 --> 00:49:40.380 align:middle line:90% doesn't say that on the slide-- 00:49:40.380 --> 00:49:44.040 align:middle line:84% the actual level of output and the predicted level of output-- 00:49:44.040 --> 00:49:45.730 align:middle line:90% model-predicted. 00:49:45.730 --> 00:49:48.060 align:middle line:84% So we ran a successful experiment. 00:49:48.060 --> 00:49:50.400 align:middle line:84% If you think about the model as an experiment, 00:49:50.400 --> 00:49:54.420 align:middle line:84% populated by the parameters we calibrate, 00:49:54.420 --> 00:49:59.130 align:middle line:84% then we are able to take this stylized model, a la Lucas, 00:49:59.130 --> 00:50:02.070 align:middle line:90% and use it to interpret reality. 00:50:02.070 --> 00:50:05.220 align:middle line:84% This is pretty close for consumption, 00:50:05.220 --> 00:50:07.870 align:middle line:84% the actual being below the model. 00:50:07.870 --> 00:50:10.380 align:middle line:84% So in this case, they're consuming a little bit less 00:50:10.380 --> 00:50:12.420 align:middle line:84% than what the model says they might, 00:50:12.420 --> 00:50:15.210 align:middle line:84% despite our attempt to fit it to the data. 00:50:15.210 --> 00:50:20.980 align:middle line:84% And this is for assets, which is very, very close. 00:50:20.980 --> 00:50:23.220 align:middle line:84% Now, these last three figures are all 00:50:23.220 --> 00:50:27.666 align:middle line:84% aggregated across all the households in a given economy. 00:50:27.666 --> 00:50:30.006 align:middle line:90% OK, so having sort of-- 00:50:30.006 --> 00:50:32.730 align:middle line:84% oh, and this is cash holdings, which 00:50:32.730 --> 00:50:36.030 align:middle line:90% is one of our weak points. 00:50:36.030 --> 00:50:37.890 align:middle line:84% And we're pretty close for a while, 00:50:37.890 --> 00:50:39.750 align:middle line:84% and then cash holdings in the data 00:50:39.750 --> 00:50:42.780 align:middle line:84% still want to be higher than in the model. 00:50:42.780 --> 00:50:45.480 align:middle line:84% God knows why they want that cash. 00:50:45.480 --> 00:50:48.180 align:middle line:90% We're just still dumbfounded. 00:50:48.180 --> 00:50:51.990 align:middle line:84% That's the way they're saving over time, in the aggregate. 00:50:51.990 --> 00:50:54.420 align:middle line:84% Accumulating claims on foreigners? 00:50:54.420 --> 00:50:56.850 align:middle line:90% Yeah, that's the paper currency. 00:50:56.850 --> 00:51:00.130 align:middle line:84% It's a claim on the central bank, and they love. 00:51:00.130 --> 00:51:01.930 align:middle line:84% It doesn't bear interest or anything. 00:51:01.930 --> 00:51:04.510 align:middle line:84% But anyway, I'll stop complaining. 00:51:04.510 --> 00:51:09.130 align:middle line:84% OK, so having done pretty well with the aggregates, 00:51:09.130 --> 00:51:11.890 align:middle line:84% one hopes we've earned the right to talk 00:51:11.890 --> 00:51:14.830 align:middle line:90% about individual experiences. 00:51:14.830 --> 00:51:18.160 align:middle line:84% Now, needless to say, there are many households in a village 00:51:18.160 --> 00:51:21.520 align:middle line:84% and a non-trivial number of households in the sample. 00:51:21.520 --> 00:51:25.720 align:middle line:84% We're fitting the aggregates via calibration. 00:51:25.720 --> 00:51:27.910 align:middle line:84% I wouldn't expect perfect performance 00:51:27.910 --> 00:51:29.530 align:middle line:90% for each and every individual. 00:51:29.530 --> 00:51:31.540 align:middle line:84% That's too much to hope for, but let me show you 00:51:31.540 --> 00:51:33.820 align:middle line:84% how well we did with household a, 00:51:33.820 --> 00:51:36.370 align:middle line:84% that labor household in Lopburi that we 00:51:36.370 --> 00:51:38.120 align:middle line:90% featured at the beginning. 00:51:38.120 --> 00:51:43.990 align:middle line:84% So this is the actual darker line, labor income. 00:51:43.990 --> 00:51:48.760 align:middle line:84% I showed you this path before, actually, in wages. 00:51:48.760 --> 00:51:51.640 align:middle line:84% And they went into livestock a bit. 00:51:51.640 --> 00:51:55.460 align:middle line:84% They went into business and borrowed to do it. 00:51:55.460 --> 00:51:58.510 align:middle line:84% This is the actual model prediction, 00:51:58.510 --> 00:52:01.960 align:middle line:84% which is a lot flatter, but it's not that far off. 00:52:01.960 --> 00:52:05.290 align:middle line:84% It matches pretty well at the beginning and the end. 00:52:05.290 --> 00:52:08.240 align:middle line:84% This is consumption, which is horrible. 00:52:08.240 --> 00:52:13.750 align:middle line:84% So for this household, the actual numbers drop 00:52:13.750 --> 00:52:17.500 align:middle line:84% lower than what the model says it ought to be, 00:52:17.500 --> 00:52:19.730 align:middle line:90% and fixed assets is closer. 00:52:19.730 --> 00:52:21.650 align:middle line:84% We do pretty well with fixed assets, 00:52:21.650 --> 00:52:24.190 align:middle line:84% including at the individual household a level. 00:52:24.190 --> 00:52:30.250 align:middle line:84% And cash, again, under and over, going up, though, for both. 00:52:30.250 --> 00:52:34.780 align:middle line:84% OK, so I'm tempted to put this in the US context, 00:52:34.780 --> 00:52:38.020 align:middle line:84% because it may make it seem more real if you 00:52:38.020 --> 00:52:40.090 align:middle line:90% haven't been in Thai villages. 00:52:40.090 --> 00:52:44.800 align:middle line:84% What's the effect on farmers of restricted exports to China, 00:52:44.800 --> 00:52:47.440 align:middle line:90% when China retaliated? 00:52:47.440 --> 00:52:50.020 align:middle line:84% They came to an agreement, but, still, China's 00:52:50.020 --> 00:52:51.940 align:middle line:90% not purchasing as much. 00:52:51.940 --> 00:52:54.520 align:middle line:84% The other way around, what about manufactured goods 00:52:54.520 --> 00:52:55.870 align:middle line:90% coming in from China? 00:52:55.870 --> 00:52:58.510 align:middle line:84% What's the effect on wage-earners 00:52:58.510 --> 00:53:02.710 align:middle line:84% that produce the thing, which is now being imported, whereas, 00:53:02.710 --> 00:53:05.590 align:middle line:84% before, had been produced domestically? 00:53:05.590 --> 00:53:08.200 align:middle line:84% But we're going to talk about those things next time, 00:53:08.200 --> 00:53:10.030 align:middle line:84% but we can already answer questions 00:53:10.030 --> 00:53:12.040 align:middle line:90% like this for Thailand. 00:53:12.040 --> 00:53:17.560 align:middle line:84% If, for example, we take the actual estimated model 00:53:17.560 --> 00:53:20.600 align:middle line:90% and run counterfactuals, OK? 00:53:20.600 --> 00:53:25.760 align:middle line:84% So suppose we impose trade frictions within the country. 00:53:25.760 --> 00:53:28.810 align:middle line:84% What are tariffs going to do, basically? 00:53:28.810 --> 00:53:30.040 align:middle line:90% They hurt trade. 00:53:30.040 --> 00:53:32.710 align:middle line:84% So if a village had been exporting 00:53:32.710 --> 00:53:36.580 align:middle line:84% a lot of agricultural goods and importing a lot of manufactured 00:53:36.580 --> 00:53:41.590 align:middle line:84% goods, as in the Northeast, then the effect of a trade friction 00:53:41.590 --> 00:53:44.170 align:middle line:84% will increase the price of the thing 00:53:44.170 --> 00:53:47.290 align:middle line:84% that they have been buying, the import good, 00:53:47.290 --> 00:53:51.400 align:middle line:84% and decrease the price of the export good. 00:53:51.400 --> 00:53:53.360 align:middle line:90% Hurt both ways. 00:53:53.360 --> 00:53:55.310 align:middle line:90% We're going to do it by 1%. 00:53:55.310 --> 00:53:58.810 align:middle line:84% So we're going to take these calibrated prices that 00:53:58.810 --> 00:54:03.130 align:middle line:84% were used to match the payments to factors and profits 00:54:03.130 --> 00:54:06.200 align:middle line:84% and now vary them and see what happens. 00:54:06.200 --> 00:54:08.470 align:middle line:84% We're also going to do a financial experiment 00:54:08.470 --> 00:54:09.460 align:middle line:90% in the model. 00:54:09.460 --> 00:54:11.900 align:middle line:84% We're going to change the interest rate. 00:54:11.900 --> 00:54:16.150 align:middle line:84% So it's like we're going to impede financial flows. 00:54:16.150 --> 00:54:19.030 align:middle line:84% So again, the effect of it always hurts. 00:54:19.030 --> 00:54:23.800 align:middle line:84% So if the village had been a net lender, for example, 00:54:23.800 --> 00:54:27.160 align:middle line:84% then the effect of the financial friction 00:54:27.160 --> 00:54:28.900 align:middle line:84% is to make the interest rate lower, 00:54:28.900 --> 00:54:31.240 align:middle line:84% because lenders don't like low interest rates. 00:54:31.240 --> 00:54:37.420 align:middle line:84% We talked about that when we did the lecture on income and price 00:54:37.420 --> 00:54:41.200 align:middle line:84% effects for households in the two-good world. 00:54:41.200 --> 00:54:42.850 align:middle line:84% Anyway, so if they're lending, they're 00:54:42.850 --> 00:54:45.880 align:middle line:84% not going to it that the interest rate goes down. 00:54:45.880 --> 00:54:49.180 align:middle line:84% And likewise, if they had been borrowing, 00:54:49.180 --> 00:54:51.670 align:middle line:84% we would have gone the other way and make 00:54:51.670 --> 00:54:53.210 align:middle line:90% the loans more expensive. 00:54:53.210 --> 00:54:57.940 align:middle line:84% So that's the effect of the way we capture in the model, 00:54:57.940 --> 00:55:03.400 align:middle line:84% in order to run an experiment, the impact of trade frictions 00:55:03.400 --> 00:55:06.970 align:middle line:84% through tariffs and/or internal frictions 00:55:06.970 --> 00:55:09.020 align:middle line:90% and financial frictions. 00:55:09.020 --> 00:55:12.310 align:middle line:84% So now you want to see what happens. 00:55:12.310 --> 00:55:14.650 align:middle line:84% This is in Lopburi, and this is the effect 00:55:14.650 --> 00:55:17.590 align:middle line:84% of that 1% increase in trade frictions, 00:55:17.590 --> 00:55:20.500 align:middle line:84% making exports give less revenue. 00:55:20.500 --> 00:55:23.110 align:middle line:90% Imports now cost more. 00:55:23.110 --> 00:55:27.820 align:middle line:84% That's going to affect the production and profits 00:55:27.820 --> 00:55:31.660 align:middle line:84% from the manufactured good and the agricultural good. 00:55:31.660 --> 00:55:35.410 align:middle line:84% And in turn, remember the algorithm from last time 00:55:35.410 --> 00:55:37.060 align:middle line:90% that I reviewed? 00:55:37.060 --> 00:55:39.940 align:middle line:84% Those price changes are going to map into-- 00:55:39.940 --> 00:55:43.870 align:middle line:84% input quantities are going to map into relative factor price 00:55:43.870 --> 00:55:45.220 align:middle line:90% movements, OK? 00:55:45.220 --> 00:55:49.620 align:middle line:84% But there's a lot of diversity in these economies. 00:55:49.620 --> 00:55:52.660 align:middle line:84% As I've been saying, households differ by their initial level 00:55:52.660 --> 00:55:53.530 align:middle line:90% of capital. 00:55:53.530 --> 00:55:56.570 align:middle line:84% They differ also by their talent. 00:55:56.570 --> 00:56:01.300 align:middle line:84% So we have capital moving on the x-axis and the average skill 00:56:01.300 --> 00:56:04.690 align:middle line:84% level, pretty high skill level, and very high skill level. 00:56:04.690 --> 00:56:08.930 align:middle line:84% And then what's getting plotted here is the gains or losses. 00:56:08.930 --> 00:56:12.340 align:middle line:90% So it's a tariff, effectively. 00:56:12.340 --> 00:56:15.190 align:middle line:90% So you might expect losses. 00:56:15.190 --> 00:56:18.250 align:middle line:84% And in fact, these average or low skilled guys 00:56:18.250 --> 00:56:22.480 align:middle line:84% do suffer losses, arguably because they're wage earners, 00:56:22.480 --> 00:56:24.250 align:middle line:90% although you can see that. 00:56:24.250 --> 00:56:29.180 align:middle line:84% And the movement in production from the capital 00:56:29.180 --> 00:56:31.480 align:middle line:84% good and the labor-intensive good 00:56:31.480 --> 00:56:33.620 align:middle line:90% was such as to hurt the wage. 00:56:33.620 --> 00:56:35.680 align:middle line:84% So remember, we did Stolper-Samuelson, 00:56:35.680 --> 00:56:37.870 align:middle line:84% where an increase in the price causes 00:56:37.870 --> 00:56:40.840 align:middle line:84% an increase in the factor price of the good 00:56:40.840 --> 00:56:43.220 align:middle line:84% that is used relatively intensively. 00:56:43.220 --> 00:56:47.320 align:middle line:84% So despite all the differences with the 2x2 from last time, 00:56:47.320 --> 00:56:50.000 align:middle line:84% that ingredient is playing a role here. 00:56:50.000 --> 00:56:52.600 align:middle line:84% So these wage-earners are hurt, initially. 00:56:52.600 --> 00:56:56.020 align:middle line:84% But then, ironically, as we move in the cross section 00:56:56.020 --> 00:56:58.420 align:middle line:84% for a higher level of capital, they 00:56:58.420 --> 00:57:01.900 align:middle line:84% were actually producers, not wage-earners. 00:57:01.900 --> 00:57:03.970 align:middle line:84% And now the lower wage helps them, 00:57:03.970 --> 00:57:07.310 align:middle line:84% because it's a lowered cost of production. 00:57:07.310 --> 00:57:08.650 align:middle line:90% So their profits go up. 00:57:08.650 --> 00:57:12.550 align:middle line:84% Now they're winning, by a little less than 1%. 00:57:12.550 --> 00:57:14.500 align:middle line:84% And if they're really, really talented, 00:57:14.500 --> 00:57:17.260 align:middle line:84% they'd been producing at a higher scale all along, 00:57:17.260 --> 00:57:21.460 align:middle line:84% so the drop in wages is even better for them. 00:57:21.460 --> 00:57:24.530 align:middle line:84% OK, I'm trying to give you some intuition here, 00:57:24.530 --> 00:57:26.530 align:middle line:84% but I decided, because time was short, 00:57:26.530 --> 00:57:30.280 align:middle line:84% I wouldn't be able to actually prove all these assertions. 00:57:30.280 --> 00:57:34.030 align:middle line:84% I'm just asking you to imagine, from the 2x2x2 00:57:34.030 --> 00:57:36.700 align:middle line:84% that we did last time, why it is we're 00:57:36.700 --> 00:57:39.460 align:middle line:90% seeing all this diversity here. 00:57:39.460 --> 00:57:42.220 align:middle line:84% The order of magnitude is, like, less than 1%. 00:57:42.220 --> 00:57:43.910 align:middle line:90% Here, it's 1%. 00:57:43.910 --> 00:57:45.020 align:middle line:90% So that's pretty small. 00:57:45.020 --> 00:57:50.020 align:middle line:84% But if you look at the financial frictions in Lopburi, 00:57:50.020 --> 00:57:53.830 align:middle line:84% in the sense that I described, you can see bigger numbers. 00:57:53.830 --> 00:57:56.410 align:middle line:84% First of all, you see winners and losers. 00:57:56.410 --> 00:58:00.220 align:middle line:84% You also see that it matters whether they were wage-earners 00:58:00.220 --> 00:58:01.750 align:middle line:90% or capitalists. 00:58:01.750 --> 00:58:07.900 align:middle line:84% But above all, you see that the high-skilled guys can benefit 00:58:07.900 --> 00:58:11.620 align:middle line:84% up to 10% from the increase in the interest rate 00:58:11.620 --> 00:58:16.330 align:middle line:84% that's associated with the financial friction, 00:58:16.330 --> 00:58:20.020 align:middle line:84% because the village had been borrowing, 00:58:20.020 --> 00:58:21.550 align:middle line:84% but some of those guys don't borrow. 00:58:21.550 --> 00:58:22.660 align:middle line:90% They lend. 00:58:22.660 --> 00:58:26.740 align:middle line:84% And likewise, there are other people here, 00:58:26.740 --> 00:58:29.260 align:middle line:84% the average-skilled guys, who are 00:58:29.260 --> 00:58:32.540 align:middle line:90% taking big losses of up to 10%. 00:58:32.540 --> 00:58:36.650 align:middle line:84% So again, one could imagine producing pictures like this. 00:58:36.650 --> 00:58:39.940 align:middle line:84% And if we go back to household a, what 00:58:39.940 --> 00:58:43.000 align:middle line:84% happens to a wage earner in the US 00:58:43.000 --> 00:58:47.200 align:middle line:84% with manufactured imports or a farmer in the Midwest 00:58:47.200 --> 00:58:49.270 align:middle line:90% with agricultural exports? 00:58:49.270 --> 00:58:54.820 align:middle line:84% What happens in Lopburi to that wage-earner under the trade 00:58:54.820 --> 00:58:57.880 align:middle line:84% friction or under the financial friction? 00:58:57.880 --> 00:58:59.080 align:middle line:90% They lose. 00:58:59.080 --> 00:59:00.820 align:middle line:90% He or she loses. 00:59:00.820 --> 00:59:02.530 align:middle line:90% That household loses. 00:59:02.530 --> 00:59:05.740 align:middle line:84% So we validated the model to try to make this realistic. 00:59:05.740 --> 00:59:07.450 align:middle line:90% It fit pretty well-- 00:59:07.450 --> 00:59:08.230 align:middle line:90% very well. 00:59:08.230 --> 00:59:10.030 align:middle line:84% At the aggregate level, it fit pretty well. 00:59:10.030 --> 00:59:11.390 align:middle line:90% At the individual level-- 00:59:11.390 --> 00:59:13.570 align:middle line:84% so if for some reason the Thai government 00:59:13.570 --> 00:59:17.470 align:middle line:84% had run this experiment or they had political issues that 00:59:17.470 --> 00:59:19.960 align:middle line:84% made the country more isolated, you 00:59:19.960 --> 00:59:24.040 align:middle line:84% could see the effect of that on individual households 00:59:24.040 --> 00:59:26.230 align:middle line:90% through the lens of the model. 00:59:26.230 --> 00:59:30.620 align:middle line:90% OK, any final questions? 00:59:30.620 --> 00:59:32.720 align:middle line:84% So again, we'll kind of do it all over again 00:59:32.720 --> 00:59:36.510 align:middle line:84% when we come to the US next time, next Thursday. 00:59:36.510 --> 00:59:42.120 align:middle line:90%