WEBVTT 00:00:00.000 --> 00:00:01.988 align:middle line:90% [SQUEAKING] 00:00:01.988 --> 00:00:03.976 align:middle line:90% [RUSTLING] 00:00:03.976 --> 00:00:06.958 align:middle line:90% [CLICKING] 00:00:06.958 --> 00:00:11.930 align:middle line:90% 00:00:11.930 --> 00:00:13.160 align:middle line:90% ROBERT TOWNSEND: All right. 00:00:13.160 --> 00:00:16.070 align:middle line:90% Let's look at this schedule. 00:00:16.070 --> 00:00:20.450 align:middle line:84% So we're going to do the second of two applications in trade 00:00:20.450 --> 00:00:26.190 align:middle line:84% today, focusing on the US and data issues. 00:00:26.190 --> 00:00:30.930 align:middle line:84% So that's where we are in terms of the schedule. 00:00:30.930 --> 00:00:33.780 align:middle line:84% In terms of the reading list, we're, 00:00:33.780 --> 00:00:36.240 align:middle line:90% again, down to lecture 15. 00:00:36.240 --> 00:00:40.320 align:middle line:84% There's two papers that are starred here. 00:00:40.320 --> 00:00:43.920 align:middle line:84% And I have a few slides on each of these-- 00:00:43.920 --> 00:00:48.150 align:middle line:84% one, this Raj Chetty et al paper having 00:00:48.150 --> 00:00:51.210 align:middle line:84% to do with the impact of COVID-19. 00:00:51.210 --> 00:00:53.970 align:middle line:84% And you'll see today how that fits in to what 00:00:53.970 --> 00:00:55.440 align:middle line:90% we're trying to accomplish. 00:00:55.440 --> 00:00:58.620 align:middle line:84% And relatedly is this paper by Scott Baker 00:00:58.620 --> 00:01:02.710 align:middle line:84% and co-authors, again, on the impact of the virus. 00:01:02.710 --> 00:01:07.170 align:middle line:84% So you know that I am very conservative about starring 00:01:07.170 --> 00:01:07.830 align:middle line:90% material. 00:01:07.830 --> 00:01:13.070 align:middle line:84% But, today, those two things are starred and given 00:01:13.070 --> 00:01:17.160 align:middle line:84% important obvious importance surrounding us of this topic. 00:01:17.160 --> 00:01:20.220 align:middle line:84% You might find them particularly interesting, 00:01:20.220 --> 00:01:22.440 align:middle line:84% and I'll introduce that work today. 00:01:22.440 --> 00:01:25.740 align:middle line:84% And also, previous lecture, there were two starred papers 00:01:25.740 --> 00:01:29.880 align:middle line:84% as well, including Dawkins, that we didn't cover in class. 00:01:29.880 --> 00:01:32.580 align:middle line:90% So that's the reading list. 00:01:32.580 --> 00:01:35.700 align:middle line:84% And then, as usual, we can review 00:01:35.700 --> 00:01:40.000 align:middle line:84% a bit what we did last time, gains and losses from trade. 00:01:40.000 --> 00:01:41.970 align:middle line:90% So it's lecture 14. 00:01:41.970 --> 00:01:44.190 align:middle line:90% So the first one I earmarked-- 00:01:44.190 --> 00:01:46.980 align:middle line:84% "There are two key features that distinguish households even 00:01:46.980 --> 00:01:48.810 align:middle line:90% within the same village." 00:01:48.810 --> 00:01:51.180 align:middle line:90% And I wrote this. 00:01:51.180 --> 00:01:54.720 align:middle line:84% "Pick one of these and be able to tell a classmate about it." 00:01:54.720 --> 00:01:58.680 align:middle line:84% Try to tell "a compelling story about how households differ." 00:01:58.680 --> 00:02:03.750 align:middle line:84% Imagine if you want to be a reporter for the news diving 00:02:03.750 --> 00:02:06.180 align:middle line:84% into a community and interviewing 00:02:06.180 --> 00:02:07.830 align:middle line:90% various types of households. 00:02:07.830 --> 00:02:10.890 align:middle line:84% So let me get volunteers on this. 00:02:10.890 --> 00:02:14.910 align:middle line:90% 00:02:14.910 --> 00:02:18.280 align:middle line:90% Two key features. 00:02:18.280 --> 00:02:20.080 align:middle line:90% We covered two key features. 00:02:20.080 --> 00:02:21.520 align:middle line:90% I'm asking you to pick one. 00:02:21.520 --> 00:02:28.410 align:middle line:90% 00:02:28.410 --> 00:02:32.250 align:middle line:84% AUDIENCE: I think the first one is talent. 00:02:32.250 --> 00:02:34.960 align:middle line:90% And the second one-- 00:02:34.960 --> 00:02:39.100 align:middle line:84% I think maybe it's a initial fixed assets, which 00:02:39.100 --> 00:02:41.020 align:middle line:90% can determine your job. 00:02:41.020 --> 00:02:45.020 align:middle line:84% For example, if your initial asset is low, 00:02:45.020 --> 00:02:46.480 align:middle line:90% maybe you are a wage earner. 00:02:46.480 --> 00:02:50.380 align:middle line:84% And if your initial fixed asset is large, 00:02:50.380 --> 00:02:53.060 align:middle line:90% you own, like, a business. 00:02:53.060 --> 00:03:00.650 align:middle line:84% And for talent, what we have discussed is that-- 00:03:00.650 --> 00:03:03.050 align:middle line:84% this is more related to the sector part. 00:03:03.050 --> 00:03:08.140 align:middle line:84% I remember that if we take all the capital and labor which 00:03:08.140 --> 00:03:11.950 align:middle line:84% is not related to talent, and we get the talent, 00:03:11.950 --> 00:03:15.200 align:middle line:84% and we have sector average talent, 00:03:15.200 --> 00:03:18.520 align:middle line:84% and if we also exclude that part, 00:03:18.520 --> 00:03:23.410 align:middle line:84% and we have household-specific talent, which will distinguish 00:03:23.410 --> 00:03:25.720 align:middle line:90% households between each other. 00:03:25.720 --> 00:03:32.470 align:middle line:84% And this talent will also be related to the jobs 00:03:32.470 --> 00:03:34.720 align:middle line:90% that the households have. 00:03:34.720 --> 00:03:40.300 align:middle line:84% And I remember there is a table in the stats which have, like, 00:03:40.300 --> 00:03:42.970 align:middle line:90% livestock and something. 00:03:42.970 --> 00:03:43.797 align:middle line:90% Yeah. 00:03:43.797 --> 00:03:45.130 align:middle line:90% ROBERT TOWNSEND: Perfect answer. 00:03:45.130 --> 00:03:47.420 align:middle line:90% That's really terrific. 00:03:47.420 --> 00:03:48.740 align:middle line:90% It's hard to add to that. 00:03:48.740 --> 00:03:49.450 align:middle line:90% Yeah. 00:03:49.450 --> 00:03:51.100 align:middle line:84% In a village, you have rich people 00:03:51.100 --> 00:03:54.640 align:middle line:84% and poor people distinguished by assets, or even 00:03:54.640 --> 00:03:56.050 align:middle line:90% real, tangible assets. 00:03:56.050 --> 00:03:58.870 align:middle line:84% And you've got people that differ in talent. 00:03:58.870 --> 00:04:01.900 align:middle line:84% The talent is something you don't see directly 00:04:01.900 --> 00:04:05.380 align:middle line:84% that you can back it out from how much they're producing, 00:04:05.380 --> 00:04:07.840 align:middle line:84% controlling for the capital and labor 00:04:07.840 --> 00:04:12.010 align:middle line:90% that they utilize in production. 00:04:12.010 --> 00:04:14.560 align:middle line:84% And so there was a big dispersion 00:04:14.560 --> 00:04:19.120 align:middle line:84% in that productivity level across different households. 00:04:19.120 --> 00:04:19.899 align:middle line:90% Great answer. 00:04:19.899 --> 00:04:20.620 align:middle line:90% Thank you. 00:04:20.620 --> 00:04:23.450 align:middle line:90% 00:04:23.450 --> 00:04:28.280 align:middle line:84% Next up, "Follow the lines of national level product 00:04:28.280 --> 00:04:32.300 align:middle line:84% and income accounts," so-called NIPA, national income 00:04:32.300 --> 00:04:33.860 align:middle line:90% and product accounts. 00:04:33.860 --> 00:04:35.450 align:middle line:84% Essentially, this question is saying, 00:04:35.450 --> 00:04:38.060 align:middle line:90% where does GDP come from? 00:04:38.060 --> 00:04:42.880 align:middle line:84% What is it, and how do we measure it? 00:04:42.880 --> 00:04:48.600 align:middle line:84% And, in particular, how can we use the household income 00:04:48.600 --> 00:04:51.420 align:middle line:90% account to get to these numbers? 00:04:51.420 --> 00:04:53.970 align:middle line:84% And today I'm in the mood just to take volunteers. 00:04:53.970 --> 00:04:58.470 align:middle line:84% So would someone like to take a crack at-- 00:04:58.470 --> 00:04:59.670 align:middle line:90% excuse me-- at that one? 00:04:59.670 --> 00:05:15.330 align:middle line:90% 00:05:15.330 --> 00:05:18.840 align:middle line:84% AUDIENCE: I know that GDP is the-- 00:05:18.840 --> 00:05:22.660 align:middle line:84% it's the sum of the final goods that are sold. 00:05:22.660 --> 00:05:25.868 align:middle line:84% So it doesn't include anything that 00:05:25.868 --> 00:05:28.410 align:middle line:84% is going to be used as an input for producing something else. 00:05:28.410 --> 00:05:33.910 align:middle line:90% 00:05:33.910 --> 00:05:36.220 align:middle line:90% ROBERT TOWNSEND: That's right. 00:05:36.220 --> 00:05:39.740 align:middle line:84% So GDP is-- we have to avoid double counting. 00:05:39.740 --> 00:05:43.450 align:middle line:84% So if a firm is producing something 00:05:43.450 --> 00:05:47.170 align:middle line:84% using capital and labor and so on, 00:05:47.170 --> 00:05:50.410 align:middle line:84% and it sells that to another firm, 00:05:50.410 --> 00:05:53.030 align:middle line:84% that becomes an input in the other firm. 00:05:53.030 --> 00:05:55.930 align:middle line:84% So the danger is double counting what 00:05:55.930 --> 00:05:58.910 align:middle line:84% is the input in one firm as the output in the other one. 00:05:58.910 --> 00:06:02.030 align:middle line:84% So what you're referring to is value added. 00:06:02.030 --> 00:06:05.770 align:middle line:84% So we take output, and we subtract off payments 00:06:05.770 --> 00:06:07.760 align:middle line:90% to factors-- 00:06:07.760 --> 00:06:10.720 align:middle line:84% sorry, subtract off intermediate goods. 00:06:10.720 --> 00:06:14.035 align:middle line:84% And the rest of it is net product being produced. 00:06:14.035 --> 00:06:18.762 align:middle line:90% 00:06:18.762 --> 00:06:23.450 align:middle line:84% Another way to put that is that, quote, "net product" 00:06:23.450 --> 00:06:25.880 align:middle line:84% at the level of a given household 00:06:25.880 --> 00:06:30.520 align:middle line:84% should add up to factor payments, 00:06:30.520 --> 00:06:40.560 align:middle line:84% so namely, wages paid, interest on capital, rent or interest 00:06:40.560 --> 00:06:44.340 align:middle line:90% if borrowed, and so on. 00:06:44.340 --> 00:06:48.530 align:middle line:84% So I think the hard part of this question, given that you-- 00:06:48.530 --> 00:06:52.490 align:middle line:84% thank you for volunteering, Armando-- 00:06:52.490 --> 00:06:55.010 align:middle line:84% is it's hard to describe in words 00:06:55.010 --> 00:06:58.700 align:middle line:84% or even remember the details of the financial accounts. 00:06:58.700 --> 00:07:02.540 align:middle line:84% We started in the lecture with the income statement 00:07:02.540 --> 00:07:07.640 align:middle line:84% of the household, which had revenues and costs. 00:07:07.640 --> 00:07:12.140 align:middle line:84% And then we added and subtracted across the various line items 00:07:12.140 --> 00:07:14.910 align:middle line:84% to produce what we called product. 00:07:14.910 --> 00:07:16.835 align:middle line:84% So you can go back and review that. 00:07:16.835 --> 00:07:20.510 align:middle line:90% 00:07:20.510 --> 00:07:22.930 align:middle line:84% And that's kind of the harder part to put into words. 00:07:22.930 --> 00:07:25.405 align:middle line:84% I'm not even doing that well myself. 00:07:25.405 --> 00:07:30.580 align:middle line:90% 00:07:30.580 --> 00:07:32.280 align:middle line:84% AUDIENCE: Can I ask a quick question? 00:07:32.280 --> 00:07:33.450 align:middle line:90% ROBERT TOWNSEND: Sure. 00:07:33.450 --> 00:07:34.470 align:middle line:90% AUDIENCE: Yeah. 00:07:34.470 --> 00:07:37.200 align:middle line:84% Can you give us some specific details 00:07:37.200 --> 00:07:42.960 align:middle line:84% about the difference between the inventories and imports? 00:07:42.960 --> 00:07:49.160 align:middle line:84% Sometimes, maybe inventories-- we calculate the inventories 00:07:49.160 --> 00:07:53.030 align:middle line:90% when we calculate the NIPA. 00:07:53.030 --> 00:07:57.120 align:middle line:90% But we subtract the inputs. 00:07:57.120 --> 00:08:04.260 align:middle line:84% But sometimes we-- maybe some inventories could be 00:08:04.260 --> 00:08:07.560 align:middle line:90% the inputs in the next period. 00:08:07.560 --> 00:08:12.420 align:middle line:84% In this period, we buy something, 00:08:12.420 --> 00:08:14.290 align:middle line:90% and we do not use them. 00:08:14.290 --> 00:08:15.810 align:middle line:90% And so they are not imports. 00:08:15.810 --> 00:08:20.220 align:middle line:84% But in the next period, they would be the inputs. 00:08:20.220 --> 00:08:25.290 align:middle line:84% Then, in the next period, when we recalculate the NIPA, 00:08:25.290 --> 00:08:28.140 align:middle line:90% we should subtract this part. 00:08:28.140 --> 00:08:32.475 align:middle line:84% We should subtract these inventories. 00:08:32.475 --> 00:08:33.350 align:middle line:90% ROBERT TOWNSEND: Yes. 00:08:33.350 --> 00:08:37.460 align:middle line:84% So you've almost given the answer. 00:08:37.460 --> 00:08:41.659 align:middle line:84% We want to keep track of the difference between buying 00:08:41.659 --> 00:08:43.559 align:middle line:90% and holding versus using. 00:08:43.559 --> 00:08:49.830 align:middle line:84% So in the productions, the income statement and so on, 00:08:49.830 --> 00:08:52.520 align:middle line:84% we have the cost of producing something. 00:08:52.520 --> 00:08:55.910 align:middle line:84% So if they take something out of inventory 00:08:55.910 --> 00:08:59.750 align:middle line:84% or buy it on the market, we value it at the cost 00:08:59.750 --> 00:09:04.330 align:middle line:84% and have it as part of the cost of production. 00:09:04.330 --> 00:09:06.900 align:middle line:84% So that's kind of the flow input. 00:09:06.900 --> 00:09:09.510 align:middle line:84% Now, if they bought something on the market 00:09:09.510 --> 00:09:12.900 align:middle line:84% but didn't utilize it yet, as in your example, 00:09:12.900 --> 00:09:16.790 align:middle line:84% then they accumulate it as inventory. 00:09:16.790 --> 00:09:20.760 align:middle line:84% And that part is an asset that appears on the balance sheet. 00:09:20.760 --> 00:09:23.400 align:middle line:90% 00:09:23.400 --> 00:09:25.610 align:middle line:84% So there is a relation here between the stocks 00:09:25.610 --> 00:09:27.620 align:middle line:90% and the flows. 00:09:27.620 --> 00:09:31.100 align:middle line:84% And indeed, in the Thai data, we have all kinds 00:09:31.100 --> 00:09:34.250 align:middle line:90% of inventory accounts. 00:09:34.250 --> 00:09:39.170 align:middle line:84% We have the rice that they produced 00:09:39.170 --> 00:09:40.950 align:middle line:90% and didn't actually sell. 00:09:40.950 --> 00:09:43.430 align:middle line:84% So we have an output inventory account. 00:09:43.430 --> 00:09:46.320 align:middle line:84% We have fertilizer that they buy, 00:09:46.320 --> 00:09:49.200 align:middle line:84% but they haven't put on the field yet. 00:09:49.200 --> 00:09:51.320 align:middle line:84% So that's an input inventory account. 00:09:51.320 --> 00:09:54.410 align:middle line:84% And those things are not trivial. 00:09:54.410 --> 00:09:56.570 align:middle line:84% But the inventory per se is a way 00:09:56.570 --> 00:09:59.165 align:middle line:90% of carrying value over time. 00:09:59.165 --> 00:10:01.750 align:middle line:90% 00:10:01.750 --> 00:10:02.670 align:middle line:90% AUDIENCE: Thank you. 00:10:02.670 --> 00:10:05.340 align:middle line:90% ROBERT TOWNSEND: You're welcome. 00:10:05.340 --> 00:10:05.840 align:middle line:90% OK. 00:10:05.840 --> 00:10:08.550 align:middle line:90% 00:10:08.550 --> 00:10:11.100 align:middle line:84% "How much do villages differ from one another 00:10:11.100 --> 00:10:14.420 align:middle line:90% in terms of production." 00:10:14.420 --> 00:10:20.990 align:middle line:84% And, in particular, "Do villages vary in the shares of payments 00:10:20.990 --> 00:10:23.620 align:middle line:90% to factors of production?" 00:10:23.620 --> 00:10:26.270 align:middle line:84% And, again, today, I'm just taking volunteers. 00:10:26.270 --> 00:10:28.825 align:middle line:90% 00:10:28.825 --> 00:10:29.450 align:middle line:90% AUDIENCE: Sure. 00:10:29.450 --> 00:10:31.860 align:middle line:90% I can take this one. 00:10:31.860 --> 00:10:35.450 align:middle line:84% So I think we had that the ones that were near the urban area-- 00:10:35.450 --> 00:10:36.950 align:middle line:90% I think that was the center-- 00:10:36.950 --> 00:10:40.040 align:middle line:84% tended to have a lot more output than the ones that 00:10:40.040 --> 00:10:43.820 align:middle line:84% were in the Northeast, the more rural and farming villages, 00:10:43.820 --> 00:10:46.130 align:middle line:84% whereas the central was more manufacturing. 00:10:46.130 --> 00:10:48.200 align:middle line:84% And we saw this difference, I guess, 00:10:48.200 --> 00:10:51.620 align:middle line:84% not only in the value of their output, 00:10:51.620 --> 00:10:54.470 align:middle line:84% but also in where their income went to. 00:10:54.470 --> 00:11:00.950 align:middle line:84% So I think we had that more went to capital in the manufacturing 00:11:00.950 --> 00:11:04.010 align:middle line:84% center areas, and more went to wages in the rural farming 00:11:04.010 --> 00:11:06.195 align:middle line:90% areas. 00:11:06.195 --> 00:11:07.070 align:middle line:90% ROBERT TOWNSEND: Yes. 00:11:07.070 --> 00:11:07.670 align:middle line:90% That's right. 00:11:07.670 --> 00:11:14.990 align:middle line:84% We had wages being relatively low where labor is abundant. 00:11:14.990 --> 00:11:19.440 align:middle line:84% That's in the agricultural northeast part of the country. 00:11:19.440 --> 00:11:21.050 align:middle line:90% And if you look at the share-- 00:11:21.050 --> 00:11:23.950 align:middle line:90% 00:11:23.950 --> 00:11:27.020 align:middle line:84% well, it's a bit tricky because there's a lot of laborers, 00:11:27.020 --> 00:11:28.190 align:middle line:90% but the wage is low. 00:11:28.190 --> 00:11:31.430 align:middle line:84% So you have to multiply price times quantity. 00:11:31.430 --> 00:11:34.730 align:middle line:84% And then you get the share of total income 00:11:34.730 --> 00:11:40.200 align:middle line:84% in the village which is attributable to labor. 00:11:40.200 --> 00:11:41.580 align:middle line:84% But you're right about the factor 00:11:41.580 --> 00:11:48.480 align:middle line:84% prices and also about the difference in levels. 00:11:48.480 --> 00:11:53.010 align:middle line:84% Those villages in regions near Bangkok 00:11:53.010 --> 00:11:56.850 align:middle line:84% have very high output levels, or GDP levels, 00:11:56.850 --> 00:11:59.130 align:middle line:84% relative to villages in the Northeast. 00:11:59.130 --> 00:12:07.275 align:middle line:90% 00:12:07.275 --> 00:12:07.775 align:middle line:90% OK. 00:12:07.775 --> 00:12:12.890 align:middle line:90% 00:12:12.890 --> 00:12:16.610 align:middle line:84% Anyway, in the Northeast, despite the lower wages, 00:12:16.610 --> 00:12:23.450 align:middle line:84% the share of labor and income was higher than 00:12:23.450 --> 00:12:24.485 align:middle line:90% in the central region. 00:12:24.485 --> 00:12:32.000 align:middle line:90% 00:12:32.000 --> 00:12:37.250 align:middle line:84% And then I'll just ask about the balance of payments. 00:12:37.250 --> 00:12:39.910 align:middle line:84% Does someone remember what the balance of payments 00:12:39.910 --> 00:12:43.405 align:middle line:90% is for a village or a country? 00:12:43.405 --> 00:12:55.030 align:middle line:90% 00:12:55.030 --> 00:12:58.692 align:middle line:84% AUDIENCE: Would that just be the various balances of-- 00:12:58.692 --> 00:13:00.400 align:middle line:84% no, that would just be the balance sheet. 00:13:00.400 --> 00:13:01.900 align:middle line:90% No. 00:13:01.900 --> 00:13:05.260 align:middle line:84% It's not [INAUDIBLE] the accumulated 00:13:05.260 --> 00:13:08.440 align:middle line:84% assets and liabilities of everyone added together? 00:13:08.440 --> 00:13:10.378 align:middle line:90% Is that it? 00:13:10.378 --> 00:13:14.430 align:middle line:84% ROBERT TOWNSEND: It has to do with trade. 00:13:14.430 --> 00:13:16.480 align:middle line:84% AUDIENCE: Is it a current account? 00:13:16.480 --> 00:13:18.660 align:middle line:90% ROBERT TOWNSEND: Yeah. 00:13:18.660 --> 00:13:21.570 align:middle line:84% AUDIENCE: Yeah, it's just the sum of the trade balance 00:13:21.570 --> 00:13:24.020 align:middle line:90% and net income transfers. 00:13:24.020 --> 00:13:25.270 align:middle line:90% ROBERT TOWNSEND: That's right. 00:13:25.270 --> 00:13:30.220 align:middle line:84% So the balance of payments is like the current account, which 00:13:30.220 --> 00:13:34.210 align:middle line:84% is the sum of the trade balance, exports minus imports, 00:13:34.210 --> 00:13:39.820 align:middle line:84% plus factor payments coming from abroad-- incoming remittances 00:13:39.820 --> 00:13:41.920 align:middle line:90% and gifts and so on. 00:13:41.920 --> 00:13:44.560 align:middle line:84% So those are the two components of the flows 00:13:44.560 --> 00:13:49.120 align:middle line:84% on the left-hand side of the so-called balance equation, 00:13:49.120 --> 00:13:50.380 align:middle line:90% balance of payments equation. 00:13:50.380 --> 00:13:52.970 align:middle line:90% 00:13:52.970 --> 00:13:56.620 align:middle line:90% What's on the right hand side? 00:13:56.620 --> 00:13:58.945 align:middle line:84% The two sides add up and are equal to each other. 00:13:58.945 --> 00:14:01.480 align:middle line:90% 00:14:01.480 --> 00:14:04.770 align:middle line:84% What happens if a village is exporting a lot 00:14:04.770 --> 00:14:08.220 align:middle line:90% and importing relatively little? 00:14:08.220 --> 00:14:09.360 align:middle line:90% Where's the money going? 00:14:09.360 --> 00:14:17.594 align:middle line:90% 00:14:17.594 --> 00:14:19.762 align:middle line:90% AUDIENCE: A financial account? 00:14:19.762 --> 00:14:21.970 align:middle line:84% ROBERT TOWNSEND: Yeah, it's in the financial account. 00:14:21.970 --> 00:14:28.690 align:middle line:84% So, essentially, they're accumulating deposits 00:14:28.690 --> 00:14:31.300 align:middle line:84% in commercial banks outside the village 00:14:31.300 --> 00:14:38.230 align:middle line:84% or accumulating currency as cash or, essentially, accumulating 00:14:38.230 --> 00:14:41.130 align:middle line:90% claims on foreigners. 00:14:41.130 --> 00:14:44.060 align:middle line:84% So if you export a lot and import little, 00:14:44.060 --> 00:14:48.390 align:middle line:90% you have an advantage. 00:14:48.390 --> 00:14:50.490 align:middle line:90% What did you do with the money? 00:14:50.490 --> 00:14:52.020 align:middle line:90% You've got to put it somewhere. 00:14:52.020 --> 00:14:54.780 align:middle line:90% 00:14:54.780 --> 00:14:59.040 align:middle line:84% Or you could have been indebted externally, 00:14:59.040 --> 00:15:02.520 align:middle line:84% and then you're going to repay some of that debt. 00:15:02.520 --> 00:15:05.360 align:middle line:84% So the claims that foreigners have on you 00:15:05.360 --> 00:15:06.530 align:middle line:90% would be diminished. 00:15:06.530 --> 00:15:09.460 align:middle line:90% 00:15:09.460 --> 00:15:11.090 align:middle line:90% All right. 00:15:11.090 --> 00:15:17.100 align:middle line:84% So we're going to revisit some of these concepts today. 00:15:17.100 --> 00:15:22.570 align:middle line:90% So let's start doing that. 00:15:22.570 --> 00:15:28.280 align:middle line:84% So lecture today, 15, data and policy in the US-- 00:15:28.280 --> 00:15:30.740 align:middle line:84% and I've mentioned this before as motivation, 00:15:30.740 --> 00:15:33.830 align:middle line:84% but here we actually have the slides-- 00:15:33.830 --> 00:15:38.480 align:middle line:84% something called the China shock from the standpoint of the US. 00:15:38.480 --> 00:15:41.390 align:middle line:84% And it has to do with international trade 00:15:41.390 --> 00:15:44.400 align:middle line:84% and the welfare gains and losses. 00:15:44.400 --> 00:15:48.920 align:middle line:84% So economists argue that trade is a good thing, 00:15:48.920 --> 00:15:50.690 align:middle line:90% that it improves welfare. 00:15:50.690 --> 00:15:53.870 align:middle line:84% But we know from two lectures now-- 00:15:53.870 --> 00:15:57.200 align:middle line:84% the two-by-two theory lecture as well as the Thai lecture-- 00:15:57.200 --> 00:16:00.620 align:middle line:84% that trade may redistribute income. 00:16:00.620 --> 00:16:03.620 align:middle line:84% The idea would be that the winners 00:16:03.620 --> 00:16:06.110 align:middle line:90% gain more than the losers lose. 00:16:06.110 --> 00:16:09.660 align:middle line:84% So, effectively, if you were to try to add it all up, 00:16:09.660 --> 00:16:13.580 align:middle line:84% it may appear somehow that free trade is a good thing. 00:16:13.580 --> 00:16:17.510 align:middle line:84% But, beneath the surface, there are winners and losers 00:16:17.510 --> 00:16:21.050 align:middle line:84% from increasing foreign competition. 00:16:21.050 --> 00:16:24.590 align:middle line:84% In some respects, this slide is an introduction to something 00:16:24.590 --> 00:16:27.740 align:middle line:84% we're going to do two or three lectures from now, which 00:16:27.740 --> 00:16:29.690 align:middle line:90% is to talk about aggregation. 00:16:29.690 --> 00:16:31.460 align:middle line:84% When we draw those pictures, we look 00:16:31.460 --> 00:16:34.610 align:middle line:84% like we have a production possibilities frontier. 00:16:34.610 --> 00:16:37.910 align:middle line:84% We have one consumer with indifference curves. 00:16:37.910 --> 00:16:42.090 align:middle line:84% It looks like, as you move from autarky to free trade, 00:16:42.090 --> 00:16:44.130 align:middle line:90% the consumer is better off. 00:16:44.130 --> 00:16:46.580 align:middle line:84% But that picture is implicitly assuming 00:16:46.580 --> 00:16:48.740 align:middle line:84% that we can aggregate households up 00:16:48.740 --> 00:16:51.680 align:middle line:84% so that it appears that there's just one representative 00:16:51.680 --> 00:16:52.850 align:middle line:90% household. 00:16:52.850 --> 00:16:57.110 align:middle line:84% In fact, beneath the surface, income is getting redistributed 00:16:57.110 --> 00:17:00.570 align:middle line:84% and some people will be losing from trade. 00:17:00.570 --> 00:17:03.260 align:middle line:84% So the movement from autarky to free trade 00:17:03.260 --> 00:17:06.410 align:middle line:84% is not necessarily Pareto improving unless you 00:17:06.410 --> 00:17:08.250 align:middle line:90% compensate the losers. 00:17:08.250 --> 00:17:11.329 align:middle line:84% So here's the Heckscher-Ohlin theorem again. 00:17:11.329 --> 00:17:15.349 align:middle line:84% "Owners of a country's abundant factors gain from trade, 00:17:15.349 --> 00:17:19.130 align:middle line:84% but owners of a country's scarce factors lose." 00:17:19.130 --> 00:17:22.069 align:middle line:84% So you may remember, if the country is 00:17:22.069 --> 00:17:26.780 align:middle line:84% abundant in one of the factors, then the price of that factor 00:17:26.780 --> 00:17:30.260 align:middle line:90% tends to be low in autarky. 00:17:30.260 --> 00:17:34.250 align:middle line:84% And that's good for producing, especially 00:17:34.250 --> 00:17:39.300 align:middle line:84% for sectors that are relatively intensive in that factor. 00:17:39.300 --> 00:17:41.840 align:middle line:84% So when you open the country up for trade, 00:17:41.840 --> 00:17:44.690 align:middle line:84% that low price will be attractive to foreigners. 00:17:44.690 --> 00:17:47.150 align:middle line:90% The country will be exporting. 00:17:47.150 --> 00:17:52.010 align:middle line:84% And that, a la Stolper-Samuelson, 00:17:52.010 --> 00:17:55.640 align:middle line:84% will raise the price of the factor that's relatively 00:17:55.640 --> 00:17:58.250 align:middle line:90% intensive in production. 00:17:58.250 --> 00:18:02.180 align:middle line:84% But the flip side is the other factor, the scarce factor, 00:18:02.180 --> 00:18:06.050 align:middle line:84% will be losing as a result of free trade. 00:18:06.050 --> 00:18:08.990 align:middle line:84% The US is abundant in skilled labor. 00:18:08.990 --> 00:18:11.840 align:middle line:84% Low-skilled labor is relatively scarce, 00:18:11.840 --> 00:18:14.610 align:middle line:84% although not nonexistent, obviously. 00:18:14.610 --> 00:18:17.930 align:middle line:84% So when the US opens up to trade with Mexico and China 00:18:17.930 --> 00:18:23.480 align:middle line:84% and so on, low-skilled workers in the US are made worse off. 00:18:23.480 --> 00:18:26.900 align:middle line:84% And I can't help but say we're in the midst of the US 00:18:26.900 --> 00:18:30.300 align:middle line:84% elections, and these are big issues. 00:18:30.300 --> 00:18:34.140 align:middle line:84% What happened in the Midwest in Michigan, Wisconsin, 00:18:34.140 --> 00:18:38.660 align:middle line:84% Pennsylvania, Ohio, and so on, referred to as the Rust Belt-- 00:18:38.660 --> 00:18:42.890 align:middle line:84% means they were doing a lot of manufacturing and so on. 00:18:42.890 --> 00:18:45.890 align:middle line:84% And the view is that many of the households 00:18:45.890 --> 00:18:50.440 align:middle line:90% were hurt by the rise of trade. 00:18:50.440 --> 00:18:56.500 align:middle line:84% China in particular emerged as a great international power 00:18:56.500 --> 00:19:00.760 align:middle line:84% and made very clear the relevance 00:19:00.760 --> 00:19:04.960 align:middle line:84% of Heckscher-Ohlin and Stolper-Samuelson and so on. 00:19:04.960 --> 00:19:08.500 align:middle line:84% Actually, the slide is a bit of a rambling 00:19:08.500 --> 00:19:11.560 align:middle line:84% because it didn't look, for a long time, 00:19:11.560 --> 00:19:14.080 align:middle line:84% as if people were being hurt from trade 00:19:14.080 --> 00:19:17.720 align:middle line:84% and people were reconsidering the consensus. 00:19:17.720 --> 00:19:21.790 align:middle line:84% But now we went right back to the original underpinnings 00:19:21.790 --> 00:19:23.980 align:middle line:90% of the theory. 00:19:23.980 --> 00:19:29.040 align:middle line:84% In particular, these trade shocks that originate in China 00:19:29.040 --> 00:19:33.300 align:middle line:84% have impacted industries, plants, and local labor markets 00:19:33.300 --> 00:19:34.680 align:middle line:90% across the US. 00:19:34.680 --> 00:19:37.180 align:middle line:84% I'll show you a picture momentarily. 00:19:37.180 --> 00:19:40.060 align:middle line:84% So we want to estimate the gains from trade. 00:19:40.060 --> 00:19:44.130 align:middle line:84% We also want to document the losses that 00:19:44.130 --> 00:19:47.070 align:middle line:90% also arise under free trade. 00:19:47.070 --> 00:19:50.250 align:middle line:84% And we want to quantify who's winning and who's losing. 00:19:50.250 --> 00:19:53.040 align:middle line:84% That was the theme of the Thai lecture 00:19:53.040 --> 00:19:57.930 align:middle line:84% last time, to distinguish labor from capital, 00:19:57.930 --> 00:20:01.560 align:middle line:84% one village from another, changing the prices, 00:20:01.560 --> 00:20:04.000 align:middle line:90% the relative prices of goods. 00:20:04.000 --> 00:20:06.150 align:middle line:90% And we did counterfactuals-- 00:20:06.150 --> 00:20:08.400 align:middle line:84% what if they hadn't become more liberal, 00:20:08.400 --> 00:20:11.800 align:middle line:84% what if we imposed a tax on imports, and so on. 00:20:11.800 --> 00:20:16.530 align:middle line:84% And we were documenting how the incidents could 00:20:16.530 --> 00:20:18.330 align:middle line:84% be good or bad, depending on wealth 00:20:18.330 --> 00:20:20.508 align:middle line:90% and talent of the households. 00:20:20.508 --> 00:20:22.050 align:middle line:84% So we're going to try to do something 00:20:22.050 --> 00:20:24.420 align:middle line:84% like that momentarily for the US. 00:20:24.420 --> 00:20:26.610 align:middle line:90% Here is a picture of China-- 00:20:26.610 --> 00:20:30.510 align:middle line:84% share of world manufacturing value added. 00:20:30.510 --> 00:20:34.140 align:middle line:84% We talked about value added a minute ago, the blue line going 00:20:34.140 --> 00:20:36.990 align:middle line:90% up steadily over the decades-- 00:20:36.990 --> 00:20:41.340 align:middle line:84% 20 years approximately-- and China's share of world 00:20:41.340 --> 00:20:42.860 align:middle line:90% manufacturing exports. 00:20:42.860 --> 00:20:44.430 align:middle line:84% So not only are they producing more, 00:20:44.430 --> 00:20:48.300 align:middle line:84% they're exporting it, also on the steady rise. 00:20:48.300 --> 00:20:51.360 align:middle line:84% And here's a picture of the US current account. 00:20:51.360 --> 00:20:54.660 align:middle line:84% We reviewed the definition of this just a moment ago, 00:20:54.660 --> 00:20:58.180 align:middle line:84% the trade balance plus incoming remittances. 00:20:58.180 --> 00:21:01.500 align:middle line:84% The current account in the US is negative 00:21:01.500 --> 00:21:05.040 align:middle line:84% and got increasingly negative at exactly the same time 00:21:05.040 --> 00:21:07.680 align:middle line:84% that China's current account was positive 00:21:07.680 --> 00:21:11.760 align:middle line:84% and increasing, peaking, say, in 2005. 00:21:11.760 --> 00:21:16.830 align:middle line:84% Now, this picture probably includes the overall US current 00:21:16.830 --> 00:21:20.040 align:middle line:84% account, not only with China, but with other countries, 00:21:20.040 --> 00:21:21.630 align:middle line:90% and likewise for China. 00:21:21.630 --> 00:21:25.410 align:middle line:84% But for all the world, it looks very bilateral here. 00:21:25.410 --> 00:21:27.660 align:middle line:84% Obviously, a big part of what's happening 00:21:27.660 --> 00:21:29.640 align:middle line:84% to the trade balance in the US has 00:21:29.640 --> 00:21:33.810 align:middle line:84% to do with China, and vice versa. 00:21:33.810 --> 00:21:37.810 align:middle line:84% And then here's a picture of the US by state. 00:21:37.810 --> 00:21:41.160 align:middle line:84% And this is from that Autor, Dorn, and Hanson reference 00:21:41.160 --> 00:21:43.300 align:middle line:90% about four slides ago. 00:21:43.300 --> 00:21:48.750 align:middle line:84% So they looked at the composition of Chinese imports 00:21:48.750 --> 00:21:53.430 align:middle line:84% into the US by sector and looked at where 00:21:53.430 --> 00:21:59.640 align:middle line:84% those goods had been produced previously by county in the US. 00:21:59.640 --> 00:22:02.430 align:middle line:84% And these dark areas are the areas 00:22:02.430 --> 00:22:04.320 align:middle line:84% that are most vulnerable in the sense 00:22:04.320 --> 00:22:07.170 align:middle line:84% that they had been producing things that 00:22:07.170 --> 00:22:09.030 align:middle line:90% are now imported from China. 00:22:09.030 --> 00:22:11.370 align:middle line:90% That doesn't mean in that state. 00:22:11.370 --> 00:22:14.190 align:middle line:90% It means for the US in general. 00:22:14.190 --> 00:22:18.300 align:middle line:84% So Chinese imports are bought by consumers 00:22:18.300 --> 00:22:20.910 align:middle line:84% all over the United States, which 00:22:20.910 --> 00:22:25.410 align:middle line:84% means that a state, like a big chunk of Tennessee, if you can 00:22:25.410 --> 00:22:29.430 align:middle line:84% see it there, which had been producing manufactured 00:22:29.430 --> 00:22:32.400 align:middle line:84% goods now being imported from China, 00:22:32.400 --> 00:22:36.370 align:middle line:84% itself would be exporting less to other states in the United 00:22:36.370 --> 00:22:38.640 align:middle line:84% States because it's not producing. 00:22:38.640 --> 00:22:42.150 align:middle line:84% Those other states are buying it directly from China. 00:22:42.150 --> 00:22:43.350 align:middle line:90% Anyway, it's kind of spotty. 00:22:43.350 --> 00:22:46.020 align:middle line:84% But you can see it's certainly not universal. 00:22:46.020 --> 00:22:49.810 align:middle line:84% And here's a measure of the current account, the trade 00:22:49.810 --> 00:22:53.680 align:middle line:90% balances plus net remittances. 00:22:53.680 --> 00:22:57.490 align:middle line:84% But this treats a state in the United States 00:22:57.490 --> 00:22:58.660 align:middle line:90% as a whole country. 00:22:58.660 --> 00:23:00.815 align:middle line:84% Or another way to put it-- it treats 00:23:00.815 --> 00:23:02.440 align:middle line:84% the states in the United States the way 00:23:02.440 --> 00:23:06.130 align:middle line:84% we were considering villages in Thailand. 00:23:06.130 --> 00:23:10.000 align:middle line:84% So you can see here that the current account, the scale 00:23:10.000 --> 00:23:12.070 align:middle line:90% that's down here, is positive. 00:23:12.070 --> 00:23:15.850 align:middle line:84% So these are net exporting states. 00:23:15.850 --> 00:23:19.780 align:middle line:84% And the red, dark orange colors are, quote, 00:23:19.780 --> 00:23:21.980 align:middle line:90% "net importing" states. 00:23:21.980 --> 00:23:25.570 align:middle line:84% So here's a review of those equations again. 00:23:25.570 --> 00:23:29.230 align:middle line:84% The current account consists of the trade balance 00:23:29.230 --> 00:23:33.780 align:middle line:84% and net incoming transfers of income. 00:23:33.780 --> 00:23:35.970 align:middle line:84% The trade balance is itself the difference 00:23:35.970 --> 00:23:38.310 align:middle line:90% between exports and imports. 00:23:38.310 --> 00:23:44.170 align:middle line:84% Net income transfers consist of private and public transfers. 00:23:44.170 --> 00:23:48.840 align:middle line:84% So in the US, for example, there's a lot of taxes 00:23:48.840 --> 00:23:51.420 align:middle line:90% and a lot of welfare programs. 00:23:51.420 --> 00:23:54.690 align:middle line:84% So, actually, a big part of the state's income 00:23:54.690 --> 00:23:59.040 align:middle line:84% is attributable to net public transfers. 00:23:59.040 --> 00:24:02.520 align:middle line:84% But that's not the only source of transfers. 00:24:02.520 --> 00:24:06.660 align:middle line:84% There are gifts and remittances across states. 00:24:06.660 --> 00:24:08.760 align:middle line:90% Now, how does one get at that? 00:24:08.760 --> 00:24:15.240 align:middle line:84% Well, there's gross state income versus gross state product. 00:24:15.240 --> 00:24:17.580 align:middle line:84% The product we talked about earlier-- 00:24:17.580 --> 00:24:21.540 align:middle line:84% 10 minutes ago-- means that it is produced in the state, 00:24:21.540 --> 00:24:24.910 align:middle line:84% hiring capital and labor and so on. 00:24:24.910 --> 00:24:28.830 align:middle line:84% But their gross state income on the income statement 00:24:28.830 --> 00:24:31.500 align:middle line:84% of households would include income 00:24:31.500 --> 00:24:36.090 align:middle line:84% that they earned in other states or that other household members 00:24:36.090 --> 00:24:40.000 align:middle line:84% earned in other states and sent back home to the home state. 00:24:40.000 --> 00:24:42.870 align:middle line:90% Fortunately, the US measures-- 00:24:42.870 --> 00:24:46.110 align:middle line:84% the Bureau of Economic Analysis measures both gross state 00:24:46.110 --> 00:24:48.000 align:middle line:90% income and gross state product. 00:24:48.000 --> 00:24:52.440 align:middle line:84% We take the difference of those, and we have net transfers. 00:24:52.440 --> 00:24:54.640 align:middle line:84% And then, finally, the current account, 00:24:54.640 --> 00:24:57.060 align:middle line:84% which is, again the trade balance and net income 00:24:57.060 --> 00:25:00.870 align:middle line:84% transfers on the left, is balanced 00:25:00.870 --> 00:25:03.580 align:middle line:90% by claims on other states. 00:25:03.580 --> 00:25:04.890 align:middle line:90% So this is what I asked about. 00:25:04.890 --> 00:25:07.740 align:middle line:84% Now, I'm going to show you a lot of data momentarily. 00:25:07.740 --> 00:25:11.400 align:middle line:84% But this equation is kind of making some assumptions. 00:25:11.400 --> 00:25:14.700 align:middle line:84% Namely, it's assuming that the financial part 00:25:14.700 --> 00:25:18.510 align:middle line:84% of the adjustment has to do with the transection of the given 00:25:18.510 --> 00:25:20.050 align:middle line:90% state with other states-- 00:25:20.050 --> 00:25:23.890 align:middle line:84% for example, trading in stocks and bonds and so on. 00:25:23.890 --> 00:25:26.040 align:middle line:90% So that's the part we want. 00:25:26.040 --> 00:25:30.690 align:middle line:84% What we don't want is state investment in the capital stock 00:25:30.690 --> 00:25:33.180 align:middle line:84% or housing investment, because those 00:25:33.180 --> 00:25:36.160 align:middle line:90% are internal within the state. 00:25:36.160 --> 00:25:39.600 align:middle line:84% So to get the change on claims in foreigners, 00:25:39.600 --> 00:25:42.270 align:middle line:84% we take the change in net worth, and then 00:25:42.270 --> 00:25:44.730 align:middle line:84% subtract off how much of that shows up 00:25:44.730 --> 00:25:47.550 align:middle line:84% as increased investment in capital and housing, 00:25:47.550 --> 00:25:53.040 align:middle line:84% and assume the rest of it is an increase in financial claims 00:25:53.040 --> 00:25:54.390 align:middle line:90% on foreign-- 00:25:54.390 --> 00:25:57.570 align:middle line:84% I want to say foreign states, on other states. 00:25:57.570 --> 00:26:00.940 align:middle line:84% And we'll see how well this works out momentarily. 00:26:00.940 --> 00:26:02.220 align:middle line:90% So what data? 00:26:02.220 --> 00:26:07.860 align:middle line:84% Well, in the US, you have to be diligent and creative to begin 00:26:07.860 --> 00:26:10.830 align:middle line:84% to measure the various components. 00:26:10.830 --> 00:26:12.480 align:middle line:90% So I won't dwell on this. 00:26:12.480 --> 00:26:15.500 align:middle line:84% But, fortunately, there is something called the commodity 00:26:15.500 --> 00:26:20.150 align:middle line:84% flow survey, which measures trade flows across states 00:26:20.150 --> 00:26:22.760 align:middle line:84% in the United States, interstate trade. 00:26:22.760 --> 00:26:24.830 align:middle line:84% The growth state income and product 00:26:24.830 --> 00:26:29.150 align:middle line:84% comes from BEA regional accounts, 00:26:29.150 --> 00:26:31.460 align:middle line:90% as does public transfers. 00:26:31.460 --> 00:26:34.250 align:middle line:84% These measures incidence of the China shock, 00:26:34.250 --> 00:26:37.580 align:middle line:84% I already showed you from Autor, Dorn, and Hanson. 00:26:37.580 --> 00:26:39.170 align:middle line:90% And we have those from David. 00:26:39.170 --> 00:26:42.830 align:middle line:84% Housing stock comes from American Community Survey. 00:26:42.830 --> 00:26:45.170 align:middle line:90% We have a price index. 00:26:45.170 --> 00:26:49.880 align:middle line:84% The socks and bonds transactions are tough to get. 00:26:49.880 --> 00:26:53.360 align:middle line:84% But we have indirect access to IRS data 00:26:53.360 --> 00:26:56.340 align:middle line:84% where people report dividends and so on. 00:26:56.340 --> 00:26:58.700 align:middle line:84% And you can kind of back out what their ownership 00:26:58.700 --> 00:27:01.790 align:middle line:84% is of the assets from how much interest earnings 00:27:01.790 --> 00:27:03.620 align:middle line:90% and so on that they report. 00:27:03.620 --> 00:27:06.170 align:middle line:84% The capital stock is also hard to get. 00:27:06.170 --> 00:27:07.490 align:middle line:90% I'm not saying much about it. 00:27:07.490 --> 00:27:11.480 align:middle line:84% Household debt is pretty well measured in the New York Fed 00:27:11.480 --> 00:27:16.280 align:middle line:84% CCP, Consumer Credit Panel, which is Equifax credit 00:27:16.280 --> 00:27:18.210 align:middle line:90% card-type data. 00:27:18.210 --> 00:27:19.880 align:middle line:90% So what is this not? 00:27:19.880 --> 00:27:21.680 align:middle line:90% It's not like Thailand. 00:27:21.680 --> 00:27:24.480 align:middle line:90% And I keep coming back to this. 00:27:24.480 --> 00:27:27.380 align:middle line:84% It's not like we're on the ground measuring 00:27:27.380 --> 00:27:30.670 align:middle line:84% the income statement, the balance sheet, cash flow and so 00:27:30.670 --> 00:27:31.170 align:middle line:90% on. 00:27:31.170 --> 00:27:33.950 align:middle line:84% Instead, we have to use the best available data 00:27:33.950 --> 00:27:36.590 align:middle line:84% in the US, which is different variables coming 00:27:36.590 --> 00:27:38.120 align:middle line:90% from different sources. 00:27:38.120 --> 00:27:41.210 align:middle line:84% Although the accounting equations should still be true, 00:27:41.210 --> 00:27:44.900 align:middle line:84% the way that we're measuring the variables is different. 00:27:44.900 --> 00:27:47.870 align:middle line:84% So what's the impact of the China shock? 00:27:47.870 --> 00:27:51.770 align:middle line:84% We use the accounting identity I just showed you, technically 00:27:51.770 --> 00:27:54.500 align:middle line:84% called "seemingly unrelated regressions," 00:27:54.500 --> 00:27:57.290 align:middle line:84% we could regress the trade balance on the China shock. 00:27:57.290 --> 00:28:00.830 align:middle line:84% We can regress gifts and remittances on the China shock. 00:28:00.830 --> 00:28:03.800 align:middle line:84% If that accounting identity is true, 00:28:03.800 --> 00:28:06.920 align:middle line:84% there is some relationship across those seemingly 00:28:06.920 --> 00:28:10.010 align:middle line:84% separately estimated regressions. 00:28:10.010 --> 00:28:14.570 align:middle line:84% And so this technique is a way to do it jointly and take 00:28:14.570 --> 00:28:17.310 align:middle line:84% into account the accounting identity. 00:28:17.310 --> 00:28:20.210 align:middle line:84% So we take, say-- let me skip down 00:28:20.210 --> 00:28:22.190 align:middle line:90% here-- a 10-year difference. 00:28:22.190 --> 00:28:24.110 align:middle line:84% You can think of this conceptually 00:28:24.110 --> 00:28:27.710 align:middle line:84% as comparing states hit by the Chinese shock, states which 00:28:27.710 --> 00:28:30.200 align:middle line:84% have not been hit by the China shock, 00:28:30.200 --> 00:28:33.710 align:middle line:84% and then comparing what's going on with their trade balance. 00:28:33.710 --> 00:28:37.190 align:middle line:84% The impact on the trade balance is negative. 00:28:37.190 --> 00:28:39.200 align:middle line:84% That is to say it's an adverse shock. 00:28:39.200 --> 00:28:41.840 align:middle line:84% The impact on the current account is negative-- 00:28:41.840 --> 00:28:43.760 align:middle line:90% also an adverse shock. 00:28:43.760 --> 00:28:47.300 align:middle line:84% Impact on transfers is positive, but not 00:28:47.300 --> 00:28:49.850 align:middle line:84% enough to overcome the difference in the trade 00:28:49.850 --> 00:28:50.810 align:middle line:90% balance. 00:28:50.810 --> 00:28:53.450 align:middle line:84% And we can break it down further. 00:28:53.450 --> 00:28:59.060 align:middle line:84% Exports-- a state specialized in manufactured goods now 00:28:59.060 --> 00:29:03.530 align:middle line:84% competing with China, exports of that state to other states 00:29:03.530 --> 00:29:04.820 align:middle line:90% goes down. 00:29:04.820 --> 00:29:08.340 align:middle line:84% Imports don't seem to move very much. 00:29:08.340 --> 00:29:10.340 align:middle line:90% They're not importing less. 00:29:10.340 --> 00:29:14.120 align:middle line:84% Public transfers, ironically, over a 10-year period, 00:29:14.120 --> 00:29:15.860 align:middle line:90% don't move much. 00:29:15.860 --> 00:29:20.780 align:middle line:84% But private transfers, quite interestingly, are positive. 00:29:20.780 --> 00:29:23.300 align:middle line:90% So this is a pretty new finding. 00:29:23.300 --> 00:29:25.020 align:middle line:90% We haven't published it yet. 00:29:25.020 --> 00:29:28.130 align:middle line:84% But it should remind you a little bit about gift 00:29:28.130 --> 00:29:29.870 align:middle line:90% giving in Thai villages. 00:29:29.870 --> 00:29:32.330 align:middle line:84% We talked about them helping each other out 00:29:32.330 --> 00:29:33.590 align:middle line:90% from various shocks. 00:29:33.590 --> 00:29:35.720 align:middle line:84% These states, obviously, quote unquote, 00:29:35.720 --> 00:29:38.060 align:middle line:84% "help each other out" in the sense 00:29:38.060 --> 00:29:41.390 align:middle line:84% that they get incoming gifts and remittances 00:29:41.390 --> 00:29:45.140 align:middle line:84% and factor payments precisely when they need it 00:29:45.140 --> 00:29:47.510 align:middle line:84% when they're in trouble because they've been hit 00:29:47.510 --> 00:29:50.130 align:middle line:90% by a long-term adverse shock. 00:29:50.130 --> 00:29:52.550 align:middle line:84% So that's kind of the, quote unquote, "left-hand side" 00:29:52.550 --> 00:29:54.060 align:middle line:90% of the equation. 00:29:54.060 --> 00:29:56.490 align:middle line:84% Now we can turn to the right-hand side, 00:29:56.490 --> 00:30:01.320 align:middle line:84% which is adjustments in physical and financial capital. 00:30:01.320 --> 00:30:03.050 align:middle line:90% And we don't find much. 00:30:03.050 --> 00:30:06.560 align:middle line:84% Housing actually went up, which is counterintuitive. 00:30:06.560 --> 00:30:09.950 align:middle line:84% And the rest of the variables are insignificant. 00:30:09.950 --> 00:30:12.980 align:middle line:84% Now, this violates the accounting identity. 00:30:12.980 --> 00:30:16.370 align:middle line:84% Something on the right-hand side has to move. 00:30:16.370 --> 00:30:18.710 align:middle line:84% But, because we've been using data 00:30:18.710 --> 00:30:20.210 align:middle line:84% from all these different sources, 00:30:20.210 --> 00:30:24.470 align:middle line:84% as well as assuming that we have a way of measuring decreased 00:30:24.470 --> 00:30:29.240 align:middle line:84% claims on other states, we're just not doing good enough. 00:30:29.240 --> 00:30:31.260 align:middle line:84% It's not a comment on our techniques. 00:30:31.260 --> 00:30:34.430 align:middle line:84% It's a comment on both the data sources as well as 00:30:34.430 --> 00:30:35.480 align:middle line:90% this assumption. 00:30:35.480 --> 00:30:38.120 align:middle line:84% And I'll keep coming back to that. 00:30:38.120 --> 00:30:42.740 align:middle line:84% In particular, how do we measure wealth and income in the US? 00:30:42.740 --> 00:30:45.200 align:middle line:84% Our measure of wealth on the right-hand side 00:30:45.200 --> 00:30:47.420 align:middle line:90% isn't working very well. 00:30:47.420 --> 00:30:51.020 align:middle line:84% It is supposed to be aggregated up at the state level. 00:30:51.020 --> 00:30:54.170 align:middle line:84% But, within a state, there's arguably a lot 00:30:54.170 --> 00:30:56.450 align:middle line:84% of heterogeneity, which is kind of why 00:30:56.450 --> 00:30:59.780 align:middle line:84% we reviewed what way villages differ from one another 00:30:59.780 --> 00:31:02.630 align:middle line:84% and households within villages differ from one another. 00:31:02.630 --> 00:31:05.000 align:middle line:84% And surely within a state in the US, 00:31:05.000 --> 00:31:07.250 align:middle line:84% you see rich people and poor people. 00:31:07.250 --> 00:31:09.300 align:middle line:84% You see a lot of different wealth groups. 00:31:09.300 --> 00:31:12.110 align:middle line:84% So the low-wealth group would correspond 00:31:12.110 --> 00:31:15.290 align:middle line:84% with low-wealth workers who rely on wages, 00:31:15.290 --> 00:31:21.050 align:middle line:84% and hopefully, on pensions, versus the higher-wealth people 00:31:21.050 --> 00:31:24.470 align:middle line:84% within a state which typically have income 00:31:24.470 --> 00:31:26.460 align:middle line:90% from their business. 00:31:26.460 --> 00:31:29.840 align:middle line:84% So when we aggregate up to the whole state, 00:31:29.840 --> 00:31:32.930 align:middle line:84% we're not able to make this distinction. 00:31:32.930 --> 00:31:35.480 align:middle line:90% So that's one potential problem. 00:31:35.480 --> 00:31:37.280 align:middle line:84% Some households' financial assets 00:31:37.280 --> 00:31:38.900 align:middle line:84% are going in one direction as they're 00:31:38.900 --> 00:31:42.410 align:middle line:84% benefited, others going in the opposite direction 00:31:42.410 --> 00:31:43.680 align:middle line:90% as they're harmed. 00:31:43.680 --> 00:31:46.610 align:middle line:84% And we take the net difference, and we're having trouble 00:31:46.610 --> 00:31:47.540 align:middle line:90% finding something. 00:31:47.540 --> 00:31:51.090 align:middle line:84% That's one plausible explanation. 00:31:51.090 --> 00:31:54.080 align:middle line:84% The other is the way the data are collected to begin with. 00:31:54.080 --> 00:31:58.340 align:middle line:84% The Federal Reserve is constructing Flow of Funds, 00:31:58.340 --> 00:32:00.950 align:middle line:84% changing the balance sheet, and using 00:32:00.950 --> 00:32:04.280 align:middle line:84% a survey called the Survey of Consumer Finances. 00:32:04.280 --> 00:32:06.350 align:middle line:84% And it does a really wonderful job. 00:32:06.350 --> 00:32:08.360 align:middle line:84% These are really important measurements. 00:32:08.360 --> 00:32:12.860 align:middle line:84% People use the Flow of Funds and the Survey of Consumer Finances 00:32:12.860 --> 00:32:16.250 align:middle line:84% all the time for US data analysis, 00:32:16.250 --> 00:32:22.070 align:middle line:84% although it's not really available at a regional level, 00:32:22.070 --> 00:32:24.290 align:middle line:84% and certainly not at the level of states. 00:32:24.290 --> 00:32:28.050 align:middle line:84% They don't gather it that way, and they won't release it. 00:32:28.050 --> 00:32:29.450 align:middle line:90% Believe me, I've tried. 00:32:29.450 --> 00:32:32.820 align:middle line:84% The flip side are the flows from the Bureau 00:32:32.820 --> 00:32:34.140 align:middle line:90% of Economic Analysis. 00:32:34.140 --> 00:32:36.360 align:middle line:84% And they do a great job, too, as I've already 00:32:36.360 --> 00:32:39.150 align:middle line:84% said, measuring income and product, 00:32:39.150 --> 00:32:43.020 align:middle line:84% not only at the state level, but at the level of municipalities 00:32:43.020 --> 00:32:45.730 align:middle line:84% and centers, districts, and so on. 00:32:45.730 --> 00:32:47.880 align:middle line:84% The problem is these different agencies 00:32:47.880 --> 00:32:50.820 align:middle line:84% are doing a good job with the thing under their domain. 00:32:50.820 --> 00:32:52.020 align:middle line:90% But they have no-- 00:32:52.020 --> 00:32:54.570 align:middle line:84% the Federal Reserve is not collecting income data, 00:32:54.570 --> 00:32:58.930 align:middle line:84% and the BEA is not collecting asset or wealth data. 00:32:58.930 --> 00:33:03.000 align:middle line:84% So the wealth changes and the income changes 00:33:03.000 --> 00:33:05.700 align:middle line:84% are not consistent with one another. 00:33:05.700 --> 00:33:10.980 align:middle line:84% They're not using the logic of the accounting identities 00:33:10.980 --> 00:33:13.530 align:middle line:90% that we've been emphasizing. 00:33:13.530 --> 00:33:15.620 align:middle line:90% What identity? 00:33:15.620 --> 00:33:19.640 align:middle line:84% Namely, when you have, say, positive savings-- 00:33:19.640 --> 00:33:22.760 align:middle line:84% that's a flow-- what are they doing with that saving? 00:33:22.760 --> 00:33:25.130 align:middle line:84% That should show up as a change in wealth, 00:33:25.130 --> 00:33:28.580 align:middle line:84% a change in real or financial assets in the balance sheet, 00:33:28.580 --> 00:33:30.680 align:middle line:84% at least adjusting for capital gains. 00:33:30.680 --> 00:33:33.540 align:middle line:84% The balance sheet can go up or down with the stock market. 00:33:33.540 --> 00:33:34.650 align:middle line:90% You don't want that. 00:33:34.650 --> 00:33:37.010 align:middle line:84% So you take out those price movements, 00:33:37.010 --> 00:33:40.220 align:middle line:84% and you're left with real balance sheet items that 00:33:40.220 --> 00:33:43.400 align:middle line:84% have to move in the sense of an accounting identity 00:33:43.400 --> 00:33:45.570 align:middle line:90% with savings. 00:33:45.570 --> 00:33:48.140 align:middle line:84% But they don't, in the data, because the data 00:33:48.140 --> 00:33:51.980 align:middle line:84% aren't measured through the lens of the accounting identity. 00:33:51.980 --> 00:33:55.220 align:middle line:84% Unlike what we showed you for Thailand, 00:33:55.220 --> 00:33:57.440 align:middle line:84% where we collected the data at the level 00:33:57.440 --> 00:33:59.600 align:middle line:84% of individual households and firms, 00:33:59.600 --> 00:34:01.560 align:middle line:84% where we had the income, the balance sheet, 00:34:01.560 --> 00:34:06.620 align:middle line:84% and so on for each one of them, the income and balance sheet 00:34:06.620 --> 00:34:09.139 align:middle line:84% have to be consistent with one another. 00:34:09.139 --> 00:34:12.170 align:middle line:84% In the US, we have these well-known surveys. 00:34:12.170 --> 00:34:15.228 align:middle line:84% I've already mentioned the Survey of Consumer Finances 00:34:15.228 --> 00:34:16.520 align:middle line:90% from the Federal Reserve Board. 00:34:16.520 --> 00:34:19.639 align:middle line:84% We have the Consumer Expenditure Survey, the Panel Study 00:34:19.639 --> 00:34:22.250 align:middle line:90% of Income Dynamics, and others. 00:34:22.250 --> 00:34:26.630 align:middle line:84% So I'm showing you a balance sheet constructed 00:34:26.630 --> 00:34:28.730 align:middle line:90% by each one of these surveys. 00:34:28.730 --> 00:34:32.210 align:middle line:84% These items should start to seem familiar. 00:34:32.210 --> 00:34:34.550 align:middle line:90% You've got current assets-- 00:34:34.550 --> 00:34:39.840 align:middle line:84% namely cash, bank accounts, certificates of deposit, 00:34:39.840 --> 00:34:40.920 align:middle line:90% and so on. 00:34:40.920 --> 00:34:43.370 align:middle line:90% So, say, the median level-- 00:34:43.370 --> 00:34:46.909 align:middle line:84% $422,000 depending on the survey-- 00:34:46.909 --> 00:34:51.980 align:middle line:84% it's only $226,000 in the survey of consumer finances. 00:34:51.980 --> 00:34:55.489 align:middle line:84% It's $600,000-- the different surveys 00:34:55.489 --> 00:34:56.760 align:middle line:90% measure different things. 00:34:56.760 --> 00:34:59.570 align:middle line:90% So that's one problem. 00:34:59.570 --> 00:35:01.730 align:middle line:84% But also, we have physical investments 00:35:01.730 --> 00:35:06.690 align:middle line:84% down here, business assets, housing assets, and so on. 00:35:06.690 --> 00:35:09.740 align:middle line:84% So this should be reminding you of the balance 00:35:09.740 --> 00:35:12.050 align:middle line:90% sheets for the Thai households. 00:35:12.050 --> 00:35:15.860 align:middle line:84% After all, the concept of a balance sheet is quite general. 00:35:15.860 --> 00:35:18.600 align:middle line:84% It's just a matter of where you're applying it. 00:35:18.600 --> 00:35:21.140 align:middle line:84% However, this is not one particular household. 00:35:21.140 --> 00:35:23.900 align:middle line:84% This is the median or average level, 00:35:23.900 --> 00:35:27.830 align:middle line:84% that sort of synthesized or synthetic balance 00:35:27.830 --> 00:35:31.730 align:middle line:84% sheet that's created for each and all of these surveys 00:35:31.730 --> 00:35:33.140 align:middle line:90% individually. 00:35:33.140 --> 00:35:36.020 align:middle line:84% And you can get the income statement. 00:35:36.020 --> 00:35:40.970 align:middle line:84% An aggregate version from the PSID, the Consumer Expenditure 00:35:40.970 --> 00:35:42.380 align:middle line:90% Survey, and so on-- 00:35:42.380 --> 00:35:49.130 align:middle line:84% median income in the US, roughly $67,000, $65,000, 00:35:49.130 --> 00:35:55.190 align:middle line:84% $83,000, $70,000 et cetera, coming from labor income, 00:35:55.190 --> 00:36:00.260 align:middle line:84% production income, other income, including interest dividends, 00:36:00.260 --> 00:36:02.400 align:middle line:90% government transfers. 00:36:02.400 --> 00:36:03.810 align:middle line:90% And we have expenditures. 00:36:03.810 --> 00:36:08.040 align:middle line:84% So expenditures would not only be depreciation and so on. 00:36:08.040 --> 00:36:10.670 align:middle line:90% But it would include taxes. 00:36:10.670 --> 00:36:12.620 align:middle line:84% And the way we do the accounting in Thailand, 00:36:12.620 --> 00:36:14.820 align:middle line:90% we have consumption over here. 00:36:14.820 --> 00:36:17.360 align:middle line:84% So, basically, that's an outflow. 00:36:17.360 --> 00:36:21.680 align:middle line:84% We had net profits taking revenue minus expenses. 00:36:21.680 --> 00:36:24.920 align:middle line:84% And from the net profits, if it were positive, 00:36:24.920 --> 00:36:28.370 align:middle line:84% we would have savings and consumption. 00:36:28.370 --> 00:36:30.110 align:middle line:90% So consumption is not included. 00:36:30.110 --> 00:36:32.090 align:middle line:90% That's why it's in red here. 00:36:32.090 --> 00:36:35.930 align:middle line:84% But it is in the Statement of Cash Flows. 00:36:35.930 --> 00:36:38.390 align:middle line:84% And here's the Statement of Cash Flows 00:36:38.390 --> 00:36:40.700 align:middle line:90% for each one of those surveys-- 00:36:40.700 --> 00:36:43.010 align:middle line:84% cash flow from production, cash flow 00:36:43.010 --> 00:36:46.580 align:middle line:84% from consumption, from investment, 00:36:46.580 --> 00:36:49.160 align:middle line:90% cash flow from financing. 00:36:49.160 --> 00:36:50.870 align:middle line:84% We introduced a cash flow statement 00:36:50.870 --> 00:36:54.290 align:middle line:84% way back when we talked about Bitcoin. 00:36:54.290 --> 00:36:57.020 align:middle line:84% We had transfer of Bitcoin accounts. 00:36:57.020 --> 00:37:00.740 align:middle line:84% I made the point that both the balance sheet and the cash flow 00:37:00.740 --> 00:37:03.680 align:middle line:84% were being represented in the same ledger. 00:37:03.680 --> 00:37:05.240 align:middle line:84% You had to own it before you could 00:37:05.240 --> 00:37:07.970 align:middle line:90% buy it-- use it to buy things. 00:37:07.970 --> 00:37:12.890 align:middle line:84% Relatedly, the change in cash flow in Bitcoin should be-- 00:37:12.890 --> 00:37:16.550 align:middle line:84% if it's outgoing, should be associated with a decrease 00:37:16.550 --> 00:37:18.680 align:middle line:90% in the ownership of the coin. 00:37:18.680 --> 00:37:21.410 align:middle line:84% Or more generally, in the accounting sense, 00:37:21.410 --> 00:37:24.500 align:middle line:84% the change in cash in the statement of cash flow 00:37:24.500 --> 00:37:27.200 align:middle line:84% should be the same as the change in cash 00:37:27.200 --> 00:37:28.670 align:middle line:90% holding in the balance sheet. 00:37:28.670 --> 00:37:30.660 align:middle line:90% It seems pretty natural. 00:37:30.660 --> 00:37:34.220 align:middle line:84% And if you collect the data from scratch, each entity, one at 00:37:34.220 --> 00:37:36.200 align:middle line:84% a time, you can run this check and make 00:37:36.200 --> 00:37:38.930 align:middle line:84% sure it's true if you're using the accounting framework. 00:37:38.930 --> 00:37:40.730 align:middle line:84% If not, then you did something wrong, 00:37:40.730 --> 00:37:42.320 align:middle line:90% and you have to start over. 00:37:42.320 --> 00:37:45.830 align:middle line:84% The US data are not collected in that way. 00:37:45.830 --> 00:37:48.170 align:middle line:90% Nobody ever ran this check. 00:37:48.170 --> 00:37:50.840 align:middle line:84% So I did it with Scott Schuh and Chris. 00:37:50.840 --> 00:37:53.930 align:middle line:84% And here you can see a big discrepancy 00:37:53.930 --> 00:37:56.210 align:middle line:84% between the change in the cash flow 00:37:56.210 --> 00:37:59.540 align:middle line:84% from the statement of cash flow and the change in cash 00:37:59.540 --> 00:38:02.120 align:middle line:84% as measured on the change in the balance sheet. 00:38:02.120 --> 00:38:04.590 align:middle line:90% How big are these numbers? 00:38:04.590 --> 00:38:08.210 align:middle line:84% Well, the percent normalized by current assets 00:38:08.210 --> 00:38:12.665 align:middle line:90% is anywhere from 16 to 37%. 00:38:12.665 --> 00:38:13.640 align:middle line:90% It's huge. 00:38:13.640 --> 00:38:15.980 align:middle line:84% In other words, something is wrong. 00:38:15.980 --> 00:38:17.900 align:middle line:84% The flows aren't measured properly, 00:38:17.900 --> 00:38:20.330 align:middle line:84% or the assets are not measured properly. 00:38:20.330 --> 00:38:21.530 align:middle line:90% Does it really matter? 00:38:21.530 --> 00:38:23.180 align:middle line:90% Can we wave our hands. 00:38:23.180 --> 00:38:26.000 align:middle line:84% Well, we tried to use the data that we have. 00:38:26.000 --> 00:38:32.310 align:middle line:84% But we could be off by as much as 37% on the balance sheet. 00:38:32.310 --> 00:38:36.320 align:middle line:84% So this is the sad state of affairs for US data. 00:38:36.320 --> 00:38:38.390 align:middle line:84% Although I must say each one of these surveys 00:38:38.390 --> 00:38:40.670 align:middle line:84% is really fantastic, it was never 00:38:40.670 --> 00:38:43.183 align:middle line:84% designed with the accounting framework in mind. 00:38:43.183 --> 00:38:44.600 align:middle line:84% And they're very good at measuring 00:38:44.600 --> 00:38:46.760 align:middle line:90% what they intended to do. 00:38:46.760 --> 00:38:50.750 align:middle line:84% But, on the other hand, they never ran this check. 00:38:50.750 --> 00:38:54.710 align:middle line:84% So the notion that the change in the balance 00:38:54.710 --> 00:38:58.340 align:middle line:84% sheet and the income statement ought to be consistent 00:38:58.340 --> 00:38:59.910 align:middle line:90% is not new. 00:38:59.910 --> 00:39:02.060 align:middle line:90% It comes with the accounts. 00:39:02.060 --> 00:39:05.180 align:middle line:84% And there is something in the US called integrated macro 00:39:05.180 --> 00:39:10.010 align:middle line:84% accounts, which is exactly what it is-- accounts integrated 00:39:10.010 --> 00:39:11.400 align:middle line:90% with each other. 00:39:11.400 --> 00:39:17.120 align:middle line:84% So this unusual cooperation between the Federal Reserve 00:39:17.120 --> 00:39:21.980 align:middle line:84% Board and the BEA tried to pool the data that each has, 00:39:21.980 --> 00:39:24.290 align:middle line:90% that each is good at collecting. 00:39:24.290 --> 00:39:26.970 align:middle line:84% But they didn't change the way they're collecting the data. 00:39:26.970 --> 00:39:29.960 align:middle line:84% So there's an errors and omission line 00:39:29.960 --> 00:39:34.850 align:middle line:84% that shows up in the accounts, which is the way they reconcile 00:39:34.850 --> 00:39:36.770 align:middle line:90% the two systems together. 00:39:36.770 --> 00:39:39.740 align:middle line:84% And, again, that error and omissions line 00:39:39.740 --> 00:39:42.020 align:middle line:90% is sometimes quite substantial. 00:39:42.020 --> 00:39:46.070 align:middle line:84% Here's a picture for these integrated financial accounts, 00:39:46.070 --> 00:39:48.635 align:middle line:84% which is really a review of what you already know. 00:39:48.635 --> 00:39:50.690 align:middle line:90% You start with production. 00:39:50.690 --> 00:39:55.070 align:middle line:84% And then you have inputs, outputs, profits, and so on, 00:39:55.070 --> 00:40:00.470 align:middle line:84% producing income and savings, if they didn't eat everything. 00:40:00.470 --> 00:40:02.810 align:middle line:84% The savings has to show up somewhere. 00:40:02.810 --> 00:40:07.070 align:middle line:84% It's showing up in physical fixed inputs, capital inputs, 00:40:07.070 --> 00:40:11.840 align:middle line:84% or it's showing up in financial inputs, net lending or asset 00:40:11.840 --> 00:40:16.410 align:middle line:84% acquisition if positive, or a decrease in borrowing, 00:40:16.410 --> 00:40:20.320 align:middle line:84% assuming that profits over here were positive. 00:40:20.320 --> 00:40:23.680 align:middle line:84% Then you adjust for capital gains and losses, 00:40:23.680 --> 00:40:26.740 align:middle line:90% as I said, on the balance sheet. 00:40:26.740 --> 00:40:30.860 align:middle line:84% And you end up with a net balance sheet position. 00:40:30.860 --> 00:40:32.800 align:middle line:84% So you see you're going from flows 00:40:32.800 --> 00:40:36.730 align:middle line:84% into changes in the stocks into the new balance sheet. 00:40:36.730 --> 00:40:37.750 align:middle line:90% You have assets. 00:40:37.750 --> 00:40:39.430 align:middle line:90% You use assets in production. 00:40:39.430 --> 00:40:41.830 align:middle line:90% You get the flow, et cetera. 00:40:41.830 --> 00:40:46.100 align:middle line:84% Inventory-- that was asked about earlier-- is an example, again, 00:40:46.100 --> 00:40:48.250 align:middle line:84% of distinguishing inputs in production 00:40:48.250 --> 00:40:50.300 align:middle line:90% from ways of savings. 00:40:50.300 --> 00:40:53.770 align:middle line:84% So this is the integrated financial accounts. 00:40:53.770 --> 00:41:00.130 align:middle line:84% Unfortunately, it is, again, not available-- 00:41:00.130 --> 00:41:03.070 align:middle line:84% it has errors and omissions, and it's not 00:41:03.070 --> 00:41:06.840 align:middle line:84% available at the state or local level. 00:41:06.840 --> 00:41:11.780 align:middle line:84% So now I want to take you through a series of examples 00:41:11.780 --> 00:41:15.980 align:middle line:84% of other things that are always in the news, 00:41:15.980 --> 00:41:17.780 align:middle line:90% that you've heard about. 00:41:17.780 --> 00:41:23.330 align:middle line:84% And I'll present those facts to you and then explain how 00:41:23.330 --> 00:41:25.520 align:middle line:84% the measurement underlying those, quote, 00:41:25.520 --> 00:41:28.620 align:middle line:84% "well-known facts" is problematic. 00:41:28.620 --> 00:41:33.920 align:middle line:84% So one, inequality-- and here I'm drawing on this paper-- 00:41:33.920 --> 00:41:36.080 align:middle line:90% Piketty, Saez, and Zucman-- 00:41:36.080 --> 00:41:38.190 align:middle line:84% and they're obviously not the only ones. 00:41:38.190 --> 00:41:42.770 align:middle line:84% The story here is the top 1% of the population 00:41:42.770 --> 00:41:46.200 align:middle line:90% own 60% of the wealth. 00:41:46.200 --> 00:41:49.250 align:middle line:84% You hear that over and over again. 00:41:49.250 --> 00:41:52.380 align:middle line:84% So where are these numbers coming from? 00:41:52.380 --> 00:41:55.070 align:middle line:84% Well, what we have is quite limited. 00:41:55.070 --> 00:42:01.860 align:middle line:84% These guys start with GDP as measured by the BEA and so on. 00:42:01.860 --> 00:42:07.460 align:middle line:84% And then they're saying, well, who actually has part of it? 00:42:07.460 --> 00:42:09.410 align:middle line:90% Who has the shares? 00:42:09.410 --> 00:42:12.330 align:middle line:90% How big are the shares of GDP? 00:42:12.330 --> 00:42:14.960 align:middle line:84% So they want to create something new, 00:42:14.960 --> 00:42:18.530 align:middle line:84% arguably, called distributional accounts, which 00:42:18.530 --> 00:42:23.090 align:middle line:84% is the way GDP is distributed in the population, 00:42:23.090 --> 00:42:27.180 align:middle line:84% saying GDP is fine for totals, for growth, and so on. 00:42:27.180 --> 00:42:30.260 align:middle line:84% But for inequality, the aggregates 00:42:30.260 --> 00:42:33.900 align:middle line:84% don't tell us the degree of inequality. 00:42:33.900 --> 00:42:39.110 align:middle line:84% So people have used different surveys to measure inequality, 00:42:39.110 --> 00:42:41.640 align:middle line:90% and there's a discrepancy. 00:42:41.640 --> 00:42:43.850 align:middle line:84% So they're saying that we really need 00:42:43.850 --> 00:42:46.370 align:middle line:90% better distributional accounts. 00:42:46.370 --> 00:42:48.590 align:middle line:84% What fraction of economic growth accrues 00:42:48.590 --> 00:42:50.870 align:middle line:84% to the bottom 50% of the population 00:42:50.870 --> 00:42:53.750 align:middle line:84% or the top 10% of the population? 00:42:53.750 --> 00:42:57.170 align:middle line:84% How much of the rise in income inequality 00:42:57.170 --> 00:43:01.820 align:middle line:84% is due to the change in labor share of output versus capital 00:43:01.820 --> 00:43:03.080 align:middle line:90% shares of output? 00:43:03.080 --> 00:43:06.230 align:middle line:84% And even within labor, what's the dispersion 00:43:06.230 --> 00:43:09.290 align:middle line:84% in labor earnings across different households 00:43:09.290 --> 00:43:10.770 align:middle line:90% with respect to capital? 00:43:10.770 --> 00:43:12.860 align:middle line:84% What's the rate of return on capital? 00:43:12.860 --> 00:43:15.320 align:middle line:84% They're complaining here that we don't have answers 00:43:15.320 --> 00:43:16.640 align:middle line:90% to these questions. 00:43:16.640 --> 00:43:18.680 align:middle line:84% Actually, we don't even have good answers 00:43:18.680 --> 00:43:22.070 align:middle line:84% to the questions about how much wealth is being redistributed 00:43:22.070 --> 00:43:24.420 align:middle line:90% through taxes and transfers. 00:43:24.420 --> 00:43:26.150 align:middle line:90% So what do they do? 00:43:26.150 --> 00:43:27.620 align:middle line:90% They make an estimate. 00:43:27.620 --> 00:43:29.240 align:middle line:90% And these guys-- 00:43:29.240 --> 00:43:31.430 align:middle line:90% Smith, Zidar, and Zwick-- 00:43:31.430 --> 00:43:35.030 align:middle line:84% complain about the estimate, and they try to do better. 00:43:35.030 --> 00:43:37.790 align:middle line:84% So why are the estimates unreliable? 00:43:37.790 --> 00:43:40.850 align:middle line:84% Because, as in the theme of this lecture, 00:43:40.850 --> 00:43:44.600 align:middle line:84% wealth and income are not independently measured. 00:43:44.600 --> 00:43:46.250 align:middle line:84% They're consistent with one another 00:43:46.250 --> 00:43:48.860 align:middle line:90% but only in a mechanical way. 00:43:48.860 --> 00:43:50.810 align:middle line:90% How wealthy is a household? 00:43:50.810 --> 00:43:52.250 align:middle line:90% We don't have the wealth. 00:43:52.250 --> 00:43:54.060 align:middle line:90% Let's look at the flows. 00:43:54.060 --> 00:43:56.720 align:middle line:84% Let's look at the income flows over time the way 00:43:56.720 --> 00:44:02.180 align:middle line:84% we did for that life cycle model for Thailand, or, for example, 00:44:02.180 --> 00:44:04.610 align:middle line:84% I showed you one slide for the US. 00:44:04.610 --> 00:44:08.030 align:middle line:84% If you had income over time and say 00:44:08.030 --> 00:44:10.250 align:middle line:84% they could put it in the bank, then you 00:44:10.250 --> 00:44:12.320 align:middle line:84% might want to discount the income 00:44:12.320 --> 00:44:16.850 align:middle line:84% and get the net present value today of future labor earnings, 00:44:16.850 --> 00:44:21.780 align:middle line:84% like human capital or flows from assets. 00:44:21.780 --> 00:44:23.900 align:middle line:84% The problem is, how do you discount it? 00:44:23.900 --> 00:44:26.960 align:middle line:90% What discount rate do you use? 00:44:26.960 --> 00:44:28.190 align:middle line:90% Piketty et al. 00:44:28.190 --> 00:44:32.480 align:middle line:84% are assuming a constant discount rate across all the households, 00:44:32.480 --> 00:44:35.360 align:middle line:84% despite the obvious heterogeneity in households-- 00:44:35.360 --> 00:44:39.620 align:middle line:84% in particular, despite the fact that some wealthier households 00:44:39.620 --> 00:44:45.110 align:middle line:84% get income from business and poorer households don't. 00:44:45.110 --> 00:44:48.320 align:middle line:84% At best, they have financial income. 00:44:48.320 --> 00:44:51.240 align:middle line:84% But it's usually wages and so on. 00:44:51.240 --> 00:44:56.120 align:middle line:84% So Zwick basically adjusts the discount rate. 00:44:56.120 --> 00:44:59.720 align:middle line:84% And they start changing the numbers. 00:44:59.720 --> 00:45:03.650 align:middle line:84% Basically, the share of wealth of the top 1% 00:45:03.650 --> 00:45:06.770 align:middle line:84% is different from what you commonly hear. 00:45:06.770 --> 00:45:10.250 align:middle line:84% The share of wealth of the top 10% of the population 00:45:10.250 --> 00:45:14.930 align:middle line:84% is a bit different from what you typically hear, 00:45:14.930 --> 00:45:18.200 align:middle line:84% because they've allowed for heterogeneity in this discount 00:45:18.200 --> 00:45:19.320 align:middle line:90% rate. 00:45:19.320 --> 00:45:22.070 align:middle line:84% So their estimates of net worth are 00:45:22.070 --> 00:45:26.070 align:middle line:84% different across these different groups. 00:45:26.070 --> 00:45:30.710 align:middle line:84% So that's one illustration of the problems with very 00:45:30.710 --> 00:45:33.480 align:middle line:90% important topical issues. 00:45:33.480 --> 00:45:36.980 align:middle line:84% So much policy work and politics is 00:45:36.980 --> 00:45:43.700 align:middle line:84% based on perceived measures of inequality that are actually 00:45:43.700 --> 00:45:46.490 align:middle line:90% not terribly reliable. 00:45:46.490 --> 00:45:48.080 align:middle line:84% I'm not saying people aren't poor, 00:45:48.080 --> 00:45:50.390 align:middle line:84% and I'm not saying there's no inequality. 00:45:50.390 --> 00:45:55.220 align:middle line:84% I just wish we had better quantitative measures of it. 00:45:55.220 --> 00:45:58.790 align:middle line:84% Second application in this terrible world 00:45:58.790 --> 00:46:00.230 align:middle line:90% we're living in-- 00:46:00.230 --> 00:46:02.070 align:middle line:90% COVID-19. 00:46:02.070 --> 00:46:05.750 align:middle line:84% So what is the impact of COVID-19 00:46:05.750 --> 00:46:10.730 align:middle line:84% on individuals, on groups, on the entire economy? 00:46:10.730 --> 00:46:14.300 align:middle line:84% Again, the problem we face in the US-- 00:46:14.300 --> 00:46:17.000 align:middle line:84% the data are coming from different sources. 00:46:17.000 --> 00:46:18.810 align:middle line:90% There's a lot of creativity. 00:46:18.810 --> 00:46:20.330 align:middle line:84% I'll show you what people have done. 00:46:20.330 --> 00:46:24.560 align:middle line:84% But the bottom line is going to be, again, the data are not 00:46:24.560 --> 00:46:28.820 align:middle line:84% integrated in the sense that the income flows are 00:46:28.820 --> 00:46:30.590 align:middle line:84% measured but are not measured in a way 00:46:30.590 --> 00:46:32.840 align:middle line:84% that would be consistent with a balance sheet. 00:46:32.840 --> 00:46:36.800 align:middle line:84% In fact, typically, we do not have both an income statement 00:46:36.800 --> 00:46:40.370 align:middle line:84% and a balance sheet for anybody in the US. 00:46:40.370 --> 00:46:42.230 align:middle line:90% So here's a great paper-- 00:46:42.230 --> 00:46:46.400 align:middle line:84% Chetty and coauthors-- which is the starred paper 00:46:46.400 --> 00:46:47.780 align:middle line:90% on the reading list. 00:46:47.780 --> 00:46:50.360 align:middle line:84% "How Did COVID-19 and Stabilization 00:46:50.360 --> 00:46:52.800 align:middle line:84% Policies Affect Spending and Employment?" 00:46:52.800 --> 00:46:55.070 align:middle line:90% And then subheader-- 00:46:55.070 --> 00:46:58.610 align:middle line:84% "A New Real-Time Economic Data Tracker." 00:46:58.610 --> 00:47:01.220 align:middle line:84% If you look at the paper, you'll see the WWW dot. 00:47:01.220 --> 00:47:03.470 align:middle line:84% You can click on it, and you'll start 00:47:03.470 --> 00:47:05.880 align:middle line:90% seeing a lot of geography. 00:47:05.880 --> 00:47:08.330 align:middle line:84% So this is a summary of what they're finding. 00:47:08.330 --> 00:47:11.270 align:middle line:84% And I should tell you, this and all the other slides 00:47:11.270 --> 00:47:15.590 align:middle line:84% are based on what was going on in the US post-March, 00:47:15.590 --> 00:47:19.160 align:middle line:84% April, the first surge, and not the surge that 00:47:19.160 --> 00:47:23.100 align:middle line:84% took place early in the summer and not the current mess. 00:47:23.100 --> 00:47:28.310 align:middle line:84% So what they find is that high-income individuals reduced 00:47:28.310 --> 00:47:31.880 align:middle line:84% spending in mid-March, particularly in areas 00:47:31.880 --> 00:47:34.580 align:middle line:84% that had high COVID infection rates, 00:47:34.580 --> 00:47:37.310 align:middle line:84% and particularly in sectors associated 00:47:37.310 --> 00:47:39.050 align:middle line:90% with physical interaction. 00:47:39.050 --> 00:47:43.370 align:middle line:84% So the reduction in spending of the individuals 00:47:43.370 --> 00:47:47.690 align:middle line:84% hurt businesses that catered to high-income individuals, 00:47:47.690 --> 00:47:52.040 align:middle line:84% namely, small businesses in affluent zip codes, 00:47:52.040 --> 00:47:53.720 align:middle line:90% like restaurants and so on. 00:47:53.720 --> 00:47:56.300 align:middle line:84% They laid off their employees, who 00:47:56.300 --> 00:47:58.820 align:middle line:84% tend to be low-income individuals, which 00:47:58.820 --> 00:48:02.180 align:middle line:84% is where we got the surge in unemployment, initially 00:48:02.180 --> 00:48:04.250 align:middle line:90% in affluent areas. 00:48:04.250 --> 00:48:06.620 align:middle line:84% You can see the local diversity here. 00:48:06.620 --> 00:48:10.580 align:middle line:84% So they go on to use an event study to then estimate 00:48:10.580 --> 00:48:12.980 align:middle line:84% the impact of policies that were supposed 00:48:12.980 --> 00:48:15.710 align:middle line:90% to mitigate the adverse impact. 00:48:15.710 --> 00:48:18.890 align:middle line:84% States varied in reopening, as you know. 00:48:18.890 --> 00:48:21.890 align:middle line:84% And, nevertheless, it didn't have a lot of impact 00:48:21.890 --> 00:48:23.180 align:middle line:90% on local employment. 00:48:23.180 --> 00:48:26.930 align:middle line:84% Stimulus payments to households were made. 00:48:26.930 --> 00:48:29.340 align:middle line:84% That did increase consumer spending, 00:48:29.340 --> 00:48:33.380 align:middle line:84% but again, had very little impact on employment 00:48:33.380 --> 00:48:35.960 align:middle line:84% because, obviously, you don't take the check 00:48:35.960 --> 00:48:41.570 align:middle line:84% and go buy something at risk in a shop that may be closed. 00:48:41.570 --> 00:48:43.340 align:middle line:84% At best, you're going to spend it on-- 00:48:43.340 --> 00:48:46.970 align:middle line:84% I'll get to the spending and saving momentarily. 00:48:46.970 --> 00:48:49.400 align:middle line:84% In particular, this Paycheck Protection Program, 00:48:49.400 --> 00:48:53.300 align:middle line:84% which is in the news again today because the BEA is under-- 00:48:53.300 --> 00:48:57.560 align:middle line:84% SBA is under scrutiny-- those loans had 00:48:57.560 --> 00:49:01.380 align:middle line:84% little impact on employment in small businesses. 00:49:01.380 --> 00:49:02.300 align:middle line:90% So I don't know. 00:49:02.300 --> 00:49:04.940 align:middle line:84% If you're macro-oriented, you might think, 00:49:04.940 --> 00:49:06.710 align:middle line:90% that's an increase in spending. 00:49:06.710 --> 00:49:08.480 align:middle line:84% It should affect aggregate demand. 00:49:08.480 --> 00:49:11.060 align:middle line:84% It should have shown up as increased employment. 00:49:11.060 --> 00:49:11.960 align:middle line:90% But it didn't. 00:49:11.960 --> 00:49:14.330 align:middle line:84% This is a study from the Philadelphia Fed. 00:49:14.330 --> 00:49:17.750 align:middle line:84% And they used payments transection data, 00:49:17.750 --> 00:49:19.250 align:middle line:90% called Fiserv. 00:49:19.250 --> 00:49:22.400 align:middle line:84% So this is related to credit/debit card transactions, 00:49:22.400 --> 00:49:24.470 align:middle line:90% but it's not Visa data. 00:49:24.470 --> 00:49:27.170 align:middle line:84% It's actually the processing company 00:49:27.170 --> 00:49:31.070 align:middle line:84% that puts devices out there so you can swipe your credit card 00:49:31.070 --> 00:49:34.130 align:middle line:84% and send a communication from the store to the bank. 00:49:34.130 --> 00:49:39.050 align:middle line:84% Those transactions data are available on Fiserv in the US. 00:49:39.050 --> 00:49:43.130 align:middle line:84% And the Philadelphia Fed study is consistent with Chetty 00:49:43.130 --> 00:49:45.710 align:middle line:84% and so on in the sense that spending on restaurants 00:49:45.710 --> 00:49:48.050 align:middle line:90% and lodgings went down. 00:49:48.050 --> 00:49:51.020 align:middle line:84% And instead, people started spending money at home 00:49:51.020 --> 00:49:53.270 align:middle line:84% on home improvement, some people. 00:49:53.270 --> 00:49:56.420 align:middle line:84% So this is, again, the story of heterogeneity. 00:49:56.420 --> 00:49:59.360 align:middle line:84% Some people were able, as essential workers, 00:49:59.360 --> 00:50:00.890 align:middle line:90% to keep working on site. 00:50:00.890 --> 00:50:04.070 align:middle line:84% Not much shows up for their spending differences. 00:50:04.070 --> 00:50:07.730 align:middle line:84% Some start working remotely, but they're still working. 00:50:07.730 --> 00:50:09.500 align:middle line:84% Their expenditures are pretty much 00:50:09.500 --> 00:50:13.400 align:middle line:84% proportional to their income, not terribly surprisingly. 00:50:13.400 --> 00:50:16.430 align:middle line:84% You have people who were laid off and furloughed. 00:50:16.430 --> 00:50:18.800 align:middle line:84% They're not getting income anymore. 00:50:18.800 --> 00:50:24.230 align:middle line:84% And the expenditures drop more for the people 00:50:24.230 --> 00:50:27.720 align:middle line:84% who had been earning the least income. 00:50:27.720 --> 00:50:29.930 align:middle line:84% So there's a big drop in expenditures 00:50:29.930 --> 00:50:33.320 align:middle line:84% to vulnerable people, people who had low income 00:50:33.320 --> 00:50:36.600 align:middle line:84% to begin with-- more on the balance sheet momentarily. 00:50:36.600 --> 00:50:39.650 align:middle line:84% Now, one potential limitation of these data 00:50:39.650 --> 00:50:43.490 align:middle line:84% is they're electronic transactions, obviously. 00:50:43.490 --> 00:50:45.260 align:middle line:90% Well, what about currency? 00:50:45.260 --> 00:50:47.300 align:middle line:84% Well, you might think, oh, come on. 00:50:47.300 --> 00:50:48.380 align:middle line:90% That's not a big deal. 00:50:48.380 --> 00:50:49.880 align:middle line:90% It is a big deal. 00:50:49.880 --> 00:50:53.690 align:middle line:84% In the US, roughly 25% of transactions 00:50:53.690 --> 00:50:57.260 align:middle line:84% are still accounted for by paper currency. 00:50:57.260 --> 00:50:59.600 align:middle line:84% And then, mystery of all mysteries, it 00:50:59.600 --> 00:51:01.400 align:middle line:90% skyrocketed upwards. 00:51:01.400 --> 00:51:04.500 align:middle line:84% And that's not just a US phenomenon. 00:51:04.500 --> 00:51:06.320 align:middle line:90% So it's kind of perplexing. 00:51:06.320 --> 00:51:09.290 align:middle line:84% You would think paper currency is pretty dirty stuff. 00:51:09.290 --> 00:51:12.530 align:middle line:84% You're exposed to the virus, handling other people's-- you 00:51:12.530 --> 00:51:15.120 align:middle line:84% eventually get their money and so on. 00:51:15.120 --> 00:51:16.550 align:middle line:84% Some people take it home at night 00:51:16.550 --> 00:51:19.640 align:middle line:84% and wash it because they don't want to be infected by it. 00:51:19.640 --> 00:51:23.630 align:middle line:84% But they seem to be doing stuff that's requiring more of it 00:51:23.630 --> 00:51:26.480 align:middle line:84% because currency outstanding, as measured by the Federal 00:51:26.480 --> 00:51:28.280 align:middle line:90% Reserve, has gone way up. 00:51:28.280 --> 00:51:31.010 align:middle line:84% And there's a shortage of coins in the US right now. 00:51:31.010 --> 00:51:32.780 align:middle line:90% So we don't know why. 00:51:32.780 --> 00:51:35.150 align:middle line:84% And it also makes the measurement 00:51:35.150 --> 00:51:37.370 align:middle line:90% under Fiserv a little trickier. 00:51:37.370 --> 00:51:40.160 align:middle line:84% This would not be true with Alipay 00:51:40.160 --> 00:51:44.390 align:middle line:84% in China because they got all the transactions data-- 00:51:44.390 --> 00:51:46.460 align:middle line:84% maybe we'll get back to that later. 00:51:46.460 --> 00:51:48.750 align:middle line:84% What about lower-income households? 00:51:48.750 --> 00:51:52.430 align:middle line:84% So here's another study by Baker and co-authors 00:51:52.430 --> 00:51:53.890 align:middle line:90% on income and liquidity. 00:51:53.890 --> 00:51:56.190 align:middle line:84% Now, this is kind of interesting. 00:51:56.190 --> 00:51:59.330 align:middle line:84% They're looking at the marginal propensity 00:51:59.330 --> 00:52:03.380 align:middle line:84% to consume out of the stimulus payments. 00:52:03.380 --> 00:52:07.490 align:middle line:84% They made friends with a not-for-profit company 00:52:07.490 --> 00:52:12.440 align:middle line:84% called SaverLife, helping working families previously 00:52:12.440 --> 00:52:14.780 align:middle line:84% for the virus, helping those families 00:52:14.780 --> 00:52:19.560 align:middle line:84% to develop better savings habits and to meet financial goals. 00:52:19.560 --> 00:52:22.760 align:middle line:84% This reminds me a lot of what we were doing in Thailand. 00:52:22.760 --> 00:52:25.430 align:middle line:84% In particular, they had individuals link there 00:52:25.430 --> 00:52:26.790 align:middle line:90% voluntarily-- 00:52:26.790 --> 00:52:30.800 align:middle line:84% if they chose to do it, link their accounts to SaverLife. 00:52:30.800 --> 00:52:34.940 align:middle line:84% So now SaverLife has access to their bank account information 00:52:34.940 --> 00:52:38.060 align:middle line:84% to see the transactions and the balance sheets. 00:52:38.060 --> 00:52:40.140 align:middle line:84% You see where I'm going with this. 00:52:40.140 --> 00:52:44.300 align:middle line:84% So now you can finally, for this special subset of households, 00:52:44.300 --> 00:52:46.040 align:middle line:90% see inflows and outflows. 00:52:46.040 --> 00:52:49.710 align:middle line:84% I guess I misspoke a minute ago when I said it's never done. 00:52:49.710 --> 00:52:53.450 align:middle line:84% This is an instance where we have a version of the income 00:52:53.450 --> 00:52:56.210 align:middle line:84% statement and a version of the balance sheet, 00:52:56.210 --> 00:52:58.580 align:middle line:84% at least for some of the line items. 00:52:58.580 --> 00:53:01.100 align:middle line:84% What happened with the stimulus payments? 00:53:01.100 --> 00:53:05.480 align:middle line:84% As soon as they get the money, they spend it within 10 days. 00:53:05.480 --> 00:53:09.620 align:middle line:84% And they spend it on food and nondurables, pay the rent-- 00:53:09.620 --> 00:53:12.110 align:middle line:90% that's in the news again lately. 00:53:12.110 --> 00:53:15.740 align:middle line:84% So you get a picture of relatively poor households. 00:53:15.740 --> 00:53:19.550 align:middle line:84% Not only that, they looked at the balance sheet 00:53:19.550 --> 00:53:23.000 align:middle line:84% to see how much, quote, "liquidity" households had. 00:53:23.000 --> 00:53:27.050 align:middle line:84% And the lower the liquid stuff in the balance sheet, 00:53:27.050 --> 00:53:30.170 align:middle line:84% the larger their marginal propensity to consume out 00:53:30.170 --> 00:53:32.360 align:middle line:90% of the stimulus payment. 00:53:32.360 --> 00:53:35.640 align:middle line:84% So this looks like households that have a pretty lean balance 00:53:35.640 --> 00:53:36.140 align:middle line:90% sheet. 00:53:36.140 --> 00:53:38.120 align:middle line:84% They do not have a lot of assets. 00:53:38.120 --> 00:53:40.380 align:middle line:84% They certainly don't have liquid assets. 00:53:40.380 --> 00:53:42.840 align:middle line:84% So then they have been saving very much. 00:53:42.840 --> 00:53:45.870 align:middle line:84% And when they get that stimulus payment, they spend it. 00:53:45.870 --> 00:53:48.770 align:middle line:84% So they must have been cutting back on expenditures, 00:53:48.770 --> 00:53:51.140 align:middle line:84% at least before they got the stimulus payment. 00:53:51.140 --> 00:53:54.110 align:middle line:84% And, likewise, they didn't seem to save it. 00:53:54.110 --> 00:53:56.900 align:middle line:84% They could have taken the income transfer and put it 00:53:56.900 --> 00:53:58.190 align:middle line:90% into their savings account. 00:53:58.190 --> 00:53:59.000 align:middle line:90% But they didn't. 00:53:59.000 --> 00:54:00.800 align:middle line:90% By and large, they spent it. 00:54:00.800 --> 00:54:03.985 align:middle line:84% Well, it makes sense if you're poor, 00:54:03.985 --> 00:54:05.360 align:middle line:84% and you don't have liquid assets, 00:54:05.360 --> 00:54:07.220 align:middle line:90% and you've lost your income. 00:54:07.220 --> 00:54:09.290 align:middle line:84% But then you come to this other study. 00:54:09.290 --> 00:54:13.850 align:middle line:84% This is JPMorgan, the bank, using their own internal data 00:54:13.850 --> 00:54:17.300 align:middle line:84% on household accounts held by JPMorgan. 00:54:17.300 --> 00:54:20.570 align:middle line:84% And they have an institute that does research. 00:54:20.570 --> 00:54:27.410 align:middle line:84% So they're reporting here on the spending of various households, 00:54:27.410 --> 00:54:29.930 align:middle line:90% including low-income households. 00:54:29.930 --> 00:54:34.160 align:middle line:84% And they find large increases in liquid assets 00:54:34.160 --> 00:54:36.830 align:middle line:84% for almost all households, including 00:54:36.830 --> 00:54:38.690 align:middle line:90% these low-income households. 00:54:38.690 --> 00:54:41.660 align:middle line:84% That's exactly the opposite of what Baker found. 00:54:41.660 --> 00:54:43.770 align:middle line:90% So it's a bit of a puzzle. 00:54:43.770 --> 00:54:45.620 align:middle line:84% Now, granted, it's a different sample. 00:54:45.620 --> 00:54:48.050 align:middle line:84% It doesn't mean that one is wrong and the other is right. 00:54:48.050 --> 00:54:50.390 align:middle line:84% But it certainly shows you that we're 00:54:50.390 --> 00:54:52.310 align:middle line:84% getting very mixed pictures here, 00:54:52.310 --> 00:54:54.800 align:middle line:84% because these are among the very few ways 00:54:54.800 --> 00:54:57.530 align:middle line:84% that you can join the stocks and flows. 00:54:57.530 --> 00:54:59.450 align:middle line:84% You can get the income and balance 00:54:59.450 --> 00:55:03.530 align:middle line:84% sheet for customers of JPMorgan if you construct it, 00:55:03.530 --> 00:55:04.490 align:middle line:90% as these guys did. 00:55:04.490 --> 00:55:06.560 align:middle line:84% And you can get it for SaverLife, 00:55:06.560 --> 00:55:10.610 align:middle line:84% which is the data that Baker and company were using. 00:55:10.610 --> 00:55:12.800 align:middle line:84% They also looked at small businesses. 00:55:12.800 --> 00:55:17.000 align:middle line:84% And they find cash balances in JPMorgan accounts 00:55:17.000 --> 00:55:21.950 align:middle line:84% of small businesses decreased when COVID arrived 00:55:21.950 --> 00:55:25.550 align:middle line:84% and decreased more in some cities than others 00:55:25.550 --> 00:55:28.820 align:middle line:84% and decreased more for certain sectors 00:55:28.820 --> 00:55:31.230 align:middle line:90% and for certain ethnicities. 00:55:31.230 --> 00:55:33.680 align:middle line:84% So, again, restaurants hit hard-- 00:55:33.680 --> 00:55:37.250 align:middle line:84% Asian-owned firms actually hugely hit, even more 00:55:37.250 --> 00:55:39.890 align:middle line:84% than Black firms, which, in turn, were hit harder 00:55:39.890 --> 00:55:42.530 align:middle line:90% than white-operated businesses. 00:55:42.530 --> 00:55:44.300 align:middle line:90% So businesses are hurting. 00:55:44.300 --> 00:55:46.430 align:middle line:84% Some people think small businesses-- 00:55:46.430 --> 00:55:49.010 align:middle line:84% small and medium enterprises are the backbone 00:55:49.010 --> 00:55:51.180 align:middle line:90% of the US and many economies. 00:55:51.180 --> 00:55:55.730 align:middle line:84% And so the adverse impact of COVID hits hard on the economy 00:55:55.730 --> 00:55:57.440 align:middle line:90% when it hits these businesses. 00:55:57.440 --> 00:56:00.170 align:middle line:84% So then you would hope this Paycheck Protection 00:56:00.170 --> 00:56:02.450 align:middle line:90% Program would have helped. 00:56:02.450 --> 00:56:06.260 align:middle line:84% So the Paycheck PPP Program passed by Congress 00:56:06.260 --> 00:56:09.530 align:middle line:84% allowed the Treasury to fund loans 00:56:09.530 --> 00:56:12.020 align:middle line:84% through the Small Business Administration. 00:56:12.020 --> 00:56:15.530 align:middle line:84% SBA loans in the US are for small businesses. 00:56:15.530 --> 00:56:17.840 align:middle line:90% They're 85% guaranteed loans. 00:56:17.840 --> 00:56:21.600 align:middle line:84% But they're initiated by commercial banks. 00:56:21.600 --> 00:56:24.800 align:middle line:84% So in order to get access to PPP, 00:56:24.800 --> 00:56:26.960 align:middle line:84% you essentially already had to have 00:56:26.960 --> 00:56:28.910 align:middle line:90% some relationship with a bank. 00:56:28.910 --> 00:56:31.640 align:middle line:84% So, in the end, who got the money? 00:56:31.640 --> 00:56:36.110 align:middle line:84% People that had bank accounts in relationship banking. 00:56:36.110 --> 00:56:38.180 align:middle line:84% This picture struck me this morning 00:56:38.180 --> 00:56:40.880 align:middle line:84% when I was reviewing before class 00:56:40.880 --> 00:56:45.710 align:middle line:84% because these red areas in March and April 00:56:45.710 --> 00:56:49.010 align:middle line:84% were pretty much low-virus areas. 00:56:49.010 --> 00:56:52.700 align:middle line:84% Of course, nowadays, today, everything is reversed. 00:56:52.700 --> 00:56:56.300 align:middle line:84% And this is where the current surge is. 00:56:56.300 --> 00:56:58.760 align:middle line:84% But it was not like this in March. 00:56:58.760 --> 00:57:01.460 align:middle line:84% Instead, it was New York, remember? 00:57:01.460 --> 00:57:04.820 align:middle line:84% Well, New York didn't get very many of these PPP loans. 00:57:04.820 --> 00:57:06.530 align:middle line:90% So the loans went-- 00:57:06.530 --> 00:57:09.800 align:middle line:90% so it's a complaint, yes. 00:57:09.800 --> 00:57:14.060 align:middle line:84% It's also a comment on the physical infrastructure. 00:57:14.060 --> 00:57:15.880 align:middle line:84% Because we don't have good measurement, 00:57:15.880 --> 00:57:19.460 align:middle line:84% we also didn't have a good way of getting purchasing power 00:57:19.460 --> 00:57:22.100 align:middle line:84% into the hands of those small businesses that 00:57:22.100 --> 00:57:23.940 align:middle line:90% really needed the liquidity. 00:57:23.940 --> 00:57:26.780 align:middle line:84% So, to bring all these pieces together, 00:57:26.780 --> 00:57:29.270 align:middle line:90% we need better data in the US. 00:57:29.270 --> 00:57:31.880 align:middle line:84% The US doesn't have surveys like what 00:57:31.880 --> 00:57:33.950 align:middle line:90% I had shown you for Thailand. 00:57:33.950 --> 00:57:36.980 align:middle line:84% And, likewise, there is payments data-- 00:57:36.980 --> 00:57:38.060 align:middle line:90% I was alluding to it-- 00:57:38.060 --> 00:57:40.910 align:middle line:84% Scott Schuh was at the Boston Fed, 00:57:40.910 --> 00:57:43.010 align:middle line:84% and they were collecting surveys. 00:57:43.010 --> 00:57:46.790 align:middle line:84% That's how we know how many people use currency in the US. 00:57:46.790 --> 00:57:49.040 align:middle line:90% So I've teamed up with Scott. 00:57:49.040 --> 00:57:52.820 align:middle line:84% And we're making plans to basically collect data 00:57:52.820 --> 00:57:54.770 align:middle line:90% like we collected in Thailand. 00:57:54.770 --> 00:57:58.010 align:middle line:84% Scott moved from Boston to West Virginia. 00:57:58.010 --> 00:58:00.140 align:middle line:84% He's at the University of West Virginia. 00:58:00.140 --> 00:58:03.140 align:middle line:84% That's a deliberate choice because that's 00:58:03.140 --> 00:58:04.490 align:middle line:90% an obvious area-- 00:58:04.490 --> 00:58:07.190 align:middle line:84% West Virginia being a relatively poor state 00:58:07.190 --> 00:58:08.180 align:middle line:90% in the United States. 00:58:08.180 --> 00:58:11.990 align:middle line:84% And that's basically what we hope 00:58:11.990 --> 00:58:13.980 align:middle line:90% will be the starting point. 00:58:13.980 --> 00:58:18.110 align:middle line:84% So we have in mind here an ambitious multiyear research 00:58:18.110 --> 00:58:21.290 align:middle line:84% program, where we're going to be collecting data 00:58:21.290 --> 00:58:23.300 align:middle line:90% according to best practices. 00:58:23.300 --> 00:58:28.160 align:middle line:84% Part of it is going to combine the Boston Fed payment surveys 00:58:28.160 --> 00:58:30.920 align:middle line:90% with my surveys from Thailand. 00:58:30.920 --> 00:58:33.320 align:middle line:84% But part of it is we're not going to collect it 00:58:33.320 --> 00:58:35.480 align:middle line:90% all with physical surveys. 00:58:35.480 --> 00:58:38.850 align:middle line:84% We want to get the electronic data. 00:58:38.850 --> 00:58:41.690 align:middle line:84% So we will go to the households and so on 00:58:41.690 --> 00:58:43.280 align:middle line:84% and try to get permission the way 00:58:43.280 --> 00:58:47.630 align:middle line:84% Baker et al. did to access the electronic data for bank 00:58:47.630 --> 00:58:50.660 align:middle line:84% account and payroll information, assuming 00:58:50.660 --> 00:58:52.580 align:middle line:90% that's the way they get paid. 00:58:52.580 --> 00:58:55.040 align:middle line:84% So we'll run a survey to see whether households 00:58:55.040 --> 00:58:58.220 align:middle line:84% are willing to share this data with us or not 00:58:58.220 --> 00:59:03.380 align:middle line:84% and figure out the best method, and then collect it, 00:59:03.380 --> 00:59:06.030 align:middle line:84% very mindful of the financial accounts. 00:59:06.030 --> 00:59:08.660 align:middle line:84% So we'll be using the framework of integrated 00:59:08.660 --> 00:59:11.900 align:middle line:84% financial accounts, linking the balance sheet to the income 00:59:11.900 --> 00:59:15.830 align:middle line:84% statement and running checks all the while 00:59:15.830 --> 00:59:19.550 align:middle line:84% so that we have as good of a survey with as few errors as 00:59:19.550 --> 00:59:20.660 align:middle line:90% possible. 00:59:20.660 --> 00:59:22.970 align:middle line:84% I wish we were a little bit further along with this. 00:59:22.970 --> 00:59:26.190 align:middle line:84% But I'll just say that if any of you are interested, 00:59:26.190 --> 00:59:30.620 align:middle line:84% let me know, because there is a lot of work to do. 00:59:30.620 --> 00:59:33.620 align:middle line:84% But, actually, nowadays, traveling 00:59:33.620 --> 00:59:35.060 align:middle line:90% isn't very easy either. 00:59:35.060 --> 00:59:38.510 align:middle line:84% But my heart is in this enterprise. 00:59:38.510 --> 00:59:41.340 align:middle line:90% So that's all for today. 00:59:41.340 --> 00:59:44.570 align:middle line:84% This is kind of the state of data in the US. 00:59:44.570 --> 00:59:49.560 align:middle line:84% And I've chosen to do this by looking at 3 topics-- 00:59:49.560 --> 00:59:54.680 align:middle line:84% one, what we did in Thailand, which is local economies-- 00:59:54.680 --> 00:59:56.450 align:middle line:90% in this case, states in the US-- 00:59:56.450 --> 00:59:59.480 align:middle line:84% looking at it through the lens of balance sheets and income 00:59:59.480 --> 01:00:01.730 align:middle line:84% statements if possible, although struggling 01:00:01.730 --> 01:00:04.250 align:middle line:84% because we don't have the data collected that way, 01:00:04.250 --> 01:00:06.980 align:middle line:84% and then looking at the impact of trade shocks 01:00:06.980 --> 01:00:09.300 align:middle line:84% and state-level balance of payments. 01:00:09.300 --> 01:00:12.500 align:middle line:84% The second was measures of inequality 01:00:12.500 --> 01:00:15.360 align:middle line:84% and how we wish they were better. 01:00:15.360 --> 01:00:19.220 align:middle line:84% And the third-- the impact of the virus, which, again, 01:00:19.220 --> 01:00:22.830 align:middle line:90% runs into the same limitations. 01:00:22.830 --> 01:00:25.570 align:middle line:90% Questions? 01:00:25.570 --> 01:00:26.070 align:middle line:90% OK. 01:00:26.070 --> 01:00:31.320 align:middle line:84% So that completes the second of the two applications of trade 01:00:31.320 --> 01:00:36.540 align:middle line:84% for Thailand in the US, back to the two-by-two-by-two. 01:00:36.540 --> 01:00:41.370 align:middle line:84% And then we'll gear up for a review of the material 01:00:41.370 --> 01:00:43.260 align:middle line:84% that we've covered since the last midterm. 01:00:43.260 --> 01:00:44.610 align:middle line:90% We'll do that next Tuesday. 01:00:44.610 --> 01:00:47.850 align:middle line:84% And you can start on Friday with Michael. 01:00:47.850 --> 01:00:50.140 align:middle line:90% So that's all for today. 01:00:50.140 --> 01:00:52.160 align:middle line:90% Thank you so much. 01:00:52.160 --> 01:01:07.360 align:middle line:90%