WEBVTT 00:00:00.000 --> 00:00:02.410 align:middle line:90% [SQUEAKING] 00:00:02.410 --> 00:00:04.338 align:middle line:90% [RUSTLING] 00:00:04.338 --> 00:00:07.712 align:middle line:90% [CLICKING] 00:00:07.712 --> 00:00:10.440 align:middle line:90% 00:00:10.440 --> 00:00:12.360 align:middle line:84% JUAN PALACIOS: So last week, before Kevin, 00:00:12.360 --> 00:00:14.735 align:middle line:84% one of the things that we did was going through the whole 00:00:14.735 --> 00:00:19.620 align:middle line:84% four-quadrant model to understand how each space, 00:00:19.620 --> 00:00:22.110 align:middle line:90% each quadrant in-- 00:00:22.110 --> 00:00:25.352 align:middle line:84% any impact in each quadrant will have an implication 00:00:25.352 --> 00:00:27.060 align:middle line:84% for the whole market equilibrium and will 00:00:27.060 --> 00:00:30.810 align:middle line:84% affect the different parts of the real estate market. 00:00:30.810 --> 00:00:34.060 align:middle line:84% Now that is one purpose of the four quadrant model. 00:00:34.060 --> 00:00:36.510 align:middle line:84% But still, there is a big question 00:00:36.510 --> 00:00:40.710 align:middle line:84% of what is actually the first trigger of climate change 00:00:40.710 --> 00:00:41.910 align:middle line:90% on the real estate industry. 00:00:41.910 --> 00:00:44.490 align:middle line:84% It's very different if the tenant 00:00:44.490 --> 00:00:46.650 align:middle line:84% is the one-- the tenants are the ones that 00:00:46.650 --> 00:00:49.950 align:middle line:84% are triggering the first response to climate change 00:00:49.950 --> 00:00:52.830 align:middle line:84% or is actually the investors and the owners 00:00:52.830 --> 00:00:55.690 align:middle line:84% and the financiers that are the ones that are triggering it. 00:00:55.690 --> 00:00:58.800 align:middle line:84% Because the responses from policymakers 00:00:58.800 --> 00:01:00.600 align:middle line:84% and the owners and the real estate 00:01:00.600 --> 00:01:02.700 align:middle line:84% investors are very different whether you 00:01:02.700 --> 00:01:04.800 align:middle line:84% want to actually compensate tenants 00:01:04.800 --> 00:01:07.840 align:middle line:84% or you want to reduce risk associated to an asset. 00:01:07.840 --> 00:01:10.840 align:middle line:84% Also, when it comes to the valuation of assets, 00:01:10.840 --> 00:01:13.510 align:middle line:84% this has very different implications because it means 00:01:13.510 --> 00:01:19.750 align:middle line:84% either you adjust the numerator of a discounted cash flow 00:01:19.750 --> 00:01:24.140 align:middle line:84% of an asset down because the exposure to climate risk-- 00:01:24.140 --> 00:01:27.370 align:middle line:84% so then you actually expect that the rents are going to go down-- 00:01:27.370 --> 00:01:29.320 align:middle line:84% or do you adjust the denominator because you 00:01:29.320 --> 00:01:31.360 align:middle line:84% need to adjust the risk associated 00:01:31.360 --> 00:01:35.030 align:middle line:84% with that asset or the access to financing of that asset? 00:01:35.030 --> 00:01:38.200 align:middle line:84% So the way that you impute climate 00:01:38.200 --> 00:01:41.570 align:middle line:84% into the model and the valuations is very different. 00:01:41.570 --> 00:01:45.400 align:middle line:90% So how it is actually being-- 00:01:45.400 --> 00:01:47.380 align:middle line:84% at the moment, how the markets are actually 00:01:47.380 --> 00:01:49.420 align:middle line:84% being impacted at the moment is something 00:01:49.420 --> 00:01:52.030 align:middle line:84% that comes down to an empirical question 00:01:52.030 --> 00:01:53.480 align:middle line:90% and to empirical papers. 00:01:53.480 --> 00:01:55.930 align:middle line:84% So one of the things that I brought here is actually 00:01:55.930 --> 00:02:01.900 align:middle line:84% seeing different papers that try to dig down into whether it is 00:02:01.900 --> 00:02:05.170 align:middle line:84% the numerator, the rental cash flows that are affected, or is 00:02:05.170 --> 00:02:06.790 align:middle line:84% actually the denominator, is actually 00:02:06.790 --> 00:02:10.060 align:middle line:84% the CapEx is the risk associated with the asset-- what 00:02:10.060 --> 00:02:10.699 align:middle line:90% is affected? 00:02:10.699 --> 00:02:14.380 align:middle line:84% So you actually understand how climate risk is actually 00:02:14.380 --> 00:02:17.590 align:middle line:84% being incorporated into markets, if at all. 00:02:17.590 --> 00:02:23.270 align:middle line:84% So there are price discounts associated with climate risk. 00:02:23.270 --> 00:02:26.422 align:middle line:84% And we had, in the green cities lecture, 00:02:26.422 --> 00:02:27.880 align:middle line:84% a whole discussion about how you do 00:02:27.880 --> 00:02:29.380 align:middle line:84% this from the empirical perspective, 00:02:29.380 --> 00:02:31.060 align:middle line:84% apple-to-apple comparison, and so forth. 00:02:31.060 --> 00:02:33.970 align:middle line:84% The way that these people try to have 00:02:33.970 --> 00:02:37.360 align:middle line:84% an apple-to-apple comparison since if you look at properties 00:02:37.360 --> 00:02:40.060 align:middle line:84% that are next to the coastline versus properties 00:02:40.060 --> 00:02:44.570 align:middle line:84% that are inland will result in a very different comparison. 00:02:44.570 --> 00:02:51.010 align:middle line:84% Because if you have a sea view, if you have these condos next 00:02:51.010 --> 00:02:55.090 align:middle line:84% to the beach and so forth, you cannot easily compare the prices 00:02:55.090 --> 00:02:58.120 align:middle line:84% of those with properties that are in midtown 00:02:58.120 --> 00:02:59.030 align:middle line:90% or in the suburbs. 00:02:59.030 --> 00:03:02.230 align:middle line:84% So the way that people are usually doing 00:03:02.230 --> 00:03:06.340 align:middle line:84% is trying to compare properties with the same level of risk, 00:03:06.340 --> 00:03:08.930 align:middle line:84% but that has a different saliency of that risk. 00:03:08.930 --> 00:03:12.850 align:middle line:84% So either properties that are having different information 00:03:12.850 --> 00:03:16.300 align:middle line:84% at the purchasing time, so they have different description 00:03:16.300 --> 00:03:19.660 align:middle line:84% in places like Zillow or Costar for commercial real estate 00:03:19.660 --> 00:03:24.520 align:middle line:84% into how the property is exposed to climate risk 00:03:24.520 --> 00:03:26.860 align:middle line:84% and how resilient it is the property 00:03:26.860 --> 00:03:28.760 align:middle line:84% because of the characteristics and so forth. 00:03:28.760 --> 00:03:31.900 align:middle line:84% So this is how they actually do this empirical work, 00:03:31.900 --> 00:03:34.900 align:middle line:84% comparing properties with the same exposure 00:03:34.900 --> 00:03:36.740 align:middle line:84% but different saliency of the risk 00:03:36.740 --> 00:03:39.940 align:middle line:84% so that you understand that when the purchase has been done, 00:03:39.940 --> 00:03:42.490 align:middle line:84% there is the property that has a more salient information 00:03:42.490 --> 00:03:46.360 align:middle line:84% about climate risk being more prone to discount 00:03:46.360 --> 00:03:49.580 align:middle line:84% those factors in the valuation in the purchase. 00:03:49.580 --> 00:03:51.550 align:middle line:84% And what you see is actually when you compare 00:03:51.550 --> 00:03:53.650 align:middle line:84% the properties with more saliency to climate risk 00:03:53.650 --> 00:03:57.700 align:middle line:84% versus not that the risk, the changes in prices 00:03:57.700 --> 00:04:01.210 align:middle line:84% are there, so around minus 3% lower 00:04:01.210 --> 00:04:03.490 align:middle line:84% prices associated with being exposed to high flooding 00:04:03.490 --> 00:04:05.140 align:middle line:84% risk, sea level rise, but there is 00:04:05.140 --> 00:04:07.630 align:middle line:84% no change in rents, which means that when you really 00:04:07.630 --> 00:04:10.450 align:middle line:84% think about it, the formula to calculate property prices, 00:04:10.450 --> 00:04:12.748 align:middle line:84% it seems that it's not coming from the NOI. 00:04:12.748 --> 00:04:14.040 align:middle line:90% It's not coming from the rents. 00:04:14.040 --> 00:04:15.550 align:middle line:84% But it's actually coming from the cap rate. 00:04:15.550 --> 00:04:17.310 align:middle line:84% That is also something that makes sense 00:04:17.310 --> 00:04:21.420 align:middle line:84% because a lot of the risk associated 00:04:21.420 --> 00:04:27.510 align:middle line:84% with being in a flood-prone area is something that 00:04:27.510 --> 00:04:29.782 align:middle line:90% is in a multi-decade horizon. 00:04:29.782 --> 00:04:32.490 align:middle line:84% It's something that is not going to happen within a lease of 5-10 00:04:32.490 --> 00:04:35.407 align:middle line:84% years, but it's usually happening 50, 60, 70 years. 00:04:35.407 --> 00:04:36.990 align:middle line:84% That's where the risk starts to become 00:04:36.990 --> 00:04:39.310 align:middle line:90% more late and more salient. 00:04:39.310 --> 00:04:43.350 align:middle line:84% So it makes sense from the conceptual side 00:04:43.350 --> 00:04:45.690 align:middle line:84% that if the property-- the prices are adjusting, 00:04:45.690 --> 00:04:47.160 align:middle line:84% it's not really coming from the NOI 00:04:47.160 --> 00:04:49.077 align:middle line:84% because there is no disruption in the business 00:04:49.077 --> 00:04:51.400 align:middle line:84% or the disruption in the amenities of those places. 00:04:51.400 --> 00:04:53.940 align:middle line:84% But it's actually the disruption into your expectations 00:04:53.940 --> 00:04:56.290 align:middle line:84% of risk associated with that down the road. 00:04:56.290 --> 00:04:58.950 align:middle line:84% You do know that, eventually, some of those properties 00:04:58.950 --> 00:05:00.960 align:middle line:84% will face some problems on destruction, 00:05:00.960 --> 00:05:05.410 align:middle line:84% and that will actually interfere the cash flow of the properties. 00:05:05.410 --> 00:05:08.800 align:middle line:84% Similar story happened when you look at the actual disasters. 00:05:08.800 --> 00:05:11.363 align:middle line:84% This is New York before and after Sandy. 00:05:11.363 --> 00:05:13.030 align:middle line:84% And what you see here is the development 00:05:13.030 --> 00:05:16.450 align:middle line:84% of the different properties, the prices 00:05:16.450 --> 00:05:18.340 align:middle line:84% in the different properties, the ones that 00:05:18.340 --> 00:05:20.350 align:middle line:90% didn't have any damage. 00:05:20.350 --> 00:05:26.710 align:middle line:84% This a slight discount around 9% reduction 00:05:26.710 --> 00:05:31.270 align:middle line:84% in prices compared to the pre-Sandy levels. 00:05:31.270 --> 00:05:34.090 align:middle line:84% That is also something that the properties with major damages 00:05:34.090 --> 00:05:37.460 align:middle line:84% end up stabilizing and also the properties with minor damages. 00:05:37.460 --> 00:05:39.400 align:middle line:84% So it seems that there is a discount there, 00:05:39.400 --> 00:05:41.320 align:middle line:84% and it's nothing related to the actual damage 00:05:41.320 --> 00:05:43.070 align:middle line:84% or the actual destruction in the property. 00:05:43.070 --> 00:05:45.190 align:middle line:84% But it's something, again, related 00:05:45.190 --> 00:05:48.250 align:middle line:84% to the update of beliefs of how those properties are going 00:05:48.250 --> 00:05:50.620 align:middle line:84% to be hit again in the future when 00:05:50.620 --> 00:05:52.250 align:middle line:90% you are making those purchases. 00:05:52.250 --> 00:05:54.280 align:middle line:84% And this is a simple graph just to illustrate, 00:05:54.280 --> 00:05:58.000 align:middle line:84% but, of course, they do all of this econometric analysis that 00:05:58.000 --> 00:06:00.400 align:middle line:84% is pressing the green cities lecture 00:06:00.400 --> 00:06:03.430 align:middle line:84% to make sure that it is nothing there that 00:06:03.430 --> 00:06:06.050 align:middle line:90% is confounding the effects. 00:06:06.050 --> 00:06:07.930 align:middle line:84% But the story is that it seems that it 00:06:07.930 --> 00:06:10.100 align:middle line:84% is-- an update in beliefs is something that 00:06:10.100 --> 00:06:11.840 align:middle line:84% is an updating beliefs that comes 00:06:11.840 --> 00:06:14.570 align:middle line:84% via the cap rates and the risk associated 00:06:14.570 --> 00:06:16.380 align:middle line:90% with those properties. 00:06:16.380 --> 00:06:18.230 align:middle line:84% And that is something that will ultimately 00:06:18.230 --> 00:06:20.630 align:middle line:90% result in reduction in prices. 00:06:20.630 --> 00:06:22.250 align:middle line:84% It is nothing that is coming down 00:06:22.250 --> 00:06:24.860 align:middle line:84% at this specific moment through the leasing 00:06:24.860 --> 00:06:26.850 align:middle line:84% and through the rents and so forth. 00:06:26.850 --> 00:06:28.430 align:middle line:84% So here, the key thing-- and that's 00:06:28.430 --> 00:06:32.240 align:middle line:84% what we are going to start talking with me, but also mainly 00:06:32.240 --> 00:06:34.388 align:middle line:84% with Siqi, is how do we actually understand better 00:06:34.388 --> 00:06:36.930 align:middle line:84% the role of those beliefs and the formation of those beliefs? 00:06:36.930 --> 00:06:42.560 align:middle line:84% Because that is the critical channels via which climate risk 00:06:42.560 --> 00:06:45.680 align:middle line:84% is actually being translated now into a specific actions as 00:06:45.680 --> 00:06:48.120 align:middle line:84% of today in the real estate market. 00:06:48.120 --> 00:06:52.120 align:middle line:84% And that is why it's important not only understanding 00:06:52.120 --> 00:06:54.250 align:middle line:84% how everything unfolds in this four quadrant model, 00:06:54.250 --> 00:06:55.810 align:middle line:84% but also understand what type of quadrant 00:06:55.810 --> 00:06:58.268 align:middle line:84% is being affected because it's very different when it comes 00:06:58.268 --> 00:07:01.720 align:middle line:84% to understanding how to tackle climate risk, whether it is 00:07:01.720 --> 00:07:04.540 align:middle line:84% the numerator, is the range, or it's actually the denominator, 00:07:04.540 --> 00:07:07.910 align:middle line:84% and it's the cap rates and the discounts associated with that. 00:07:07.910 --> 00:07:13.922 align:middle line:84% Now this is-- the discount rates are especially important when 00:07:13.922 --> 00:07:16.630 align:middle line:84% you also start thinking not only about the value of those assets, 00:07:16.630 --> 00:07:18.740 align:middle line:84% but also investment assessment in those assets. 00:07:18.740 --> 00:07:20.710 align:middle line:84% Because when you are thinking about investing 00:07:20.710 --> 00:07:22.990 align:middle line:84% in climate resiliency measures, what you are doing 00:07:22.990 --> 00:07:26.920 align:middle line:84% is, again, similar to the energy efficiency story, 00:07:26.920 --> 00:07:30.550 align:middle line:84% you are investing upfront to avoid future risk 00:07:30.550 --> 00:07:34.970 align:middle line:84% and avoid future penalties or future stranded asset problems. 00:07:34.970 --> 00:07:38.230 align:middle line:90% So now the difference with-- 00:07:38.230 --> 00:07:41.140 align:middle line:84% a lot of the difference with climate risk 00:07:41.140 --> 00:07:44.710 align:middle line:84% is that it's even more important than discounting 00:07:44.710 --> 00:07:48.010 align:middle line:84% that with a lot of the energy efficiency and transition risk 00:07:48.010 --> 00:07:50.540 align:middle line:84% issues because with transition risk, with energy efficiency, 00:07:50.540 --> 00:07:53.000 align:middle line:84% we are looking at things like Local Law 97 00:07:53.000 --> 00:07:55.880 align:middle line:84% that is happening next year and is 00:07:55.880 --> 00:07:59.000 align:middle line:84% going to have even more penalties over the next decade 00:07:59.000 --> 00:07:59.630 align:middle line:90% or two. 00:07:59.630 --> 00:08:01.880 align:middle line:84% But climate risks-- we are talking about probabilities 00:08:01.880 --> 00:08:04.340 align:middle line:84% of bad stuff to happen in the next 50 years, 00:08:04.340 --> 00:08:05.430 align:middle line:90% in the next 100 years. 00:08:05.430 --> 00:08:07.670 align:middle line:84% So any minor change in the discount rate 00:08:07.670 --> 00:08:10.250 align:middle line:84% is going to have major differences 00:08:10.250 --> 00:08:12.350 align:middle line:84% into how the expected damages that 00:08:12.350 --> 00:08:13.910 align:middle line:84% are going to happen in the future 00:08:13.910 --> 00:08:17.420 align:middle line:84% are going to impact the valuation today 00:08:17.420 --> 00:08:19.160 align:middle line:84% because you are compounding things 00:08:19.160 --> 00:08:23.880 align:middle line:84% that are happening 10, 20, 30, 40, 50, 100 years from now. 00:08:23.880 --> 00:08:27.920 align:middle line:84% So any minor tweak there is going to bring these guys down 00:08:27.920 --> 00:08:30.020 align:middle line:90% a lot or up a lot. 00:08:30.020 --> 00:08:32.570 align:middle line:84% So there is a lot of discussion about what will 00:08:32.570 --> 00:08:34.110 align:middle line:90% be the right discount rate. 00:08:34.110 --> 00:08:37.919 align:middle line:84% And I'm going to actually show you a very clear example. 00:08:37.919 --> 00:08:39.980 align:middle line:84% Around the corner, there is a report 00:08:39.980 --> 00:08:42.980 align:middle line:84% about how to think about resiliency investments 00:08:42.980 --> 00:08:46.590 align:middle line:84% at the district level in Boston seaport by the city of Boston. 00:08:46.590 --> 00:08:50.000 align:middle line:84% So if you see the numbers, you see that when-- 00:08:50.000 --> 00:08:53.750 align:middle line:84% for the different predictions of sea level rise 00:08:53.750 --> 00:08:55.892 align:middle line:90% by the different decades-- 00:08:55.892 --> 00:08:57.350 align:middle line:84% so you see that now, at the moment, 00:08:57.350 --> 00:08:58.860 align:middle line:90% there is no sea level rise. 00:08:58.860 --> 00:09:00.030 align:middle line:90% This is the baseline. 00:09:00.030 --> 00:09:03.620 align:middle line:84% Then you see it's expected to be 9 inches by 2030, 00:09:03.620 --> 00:09:12.140 align:middle line:84% 21 inches by 2050, 40 inches higher by 2070. 00:09:12.140 --> 00:09:14.248 align:middle line:84% Those are the predictions of the city of Boston. 00:09:14.248 --> 00:09:16.790 align:middle line:84% And these are the damages that you have-- so different number 00:09:16.790 --> 00:09:19.207 align:middle line:84% of people that are affected, different number of buildings 00:09:19.207 --> 00:09:22.580 align:middle line:84% that are affected, and what is the value of those buildings, 00:09:22.580 --> 00:09:27.560 align:middle line:84% and what are the cost in terms of property damages and lives 00:09:27.560 --> 00:09:28.620 align:middle line:90% being disrupted? 00:09:28.620 --> 00:09:31.770 align:middle line:84% It goes up also as the sea level goes up. 00:09:31.770 --> 00:09:34.730 align:middle line:84% So if there is no resiliency interventions, 00:09:34.730 --> 00:09:38.510 align:middle line:84% this is what the damage will be-- 00:09:38.510 --> 00:09:41.330 align:middle line:84% will start around $140 million as of now, all the way 00:09:41.330 --> 00:09:43.710 align:middle line:90% to $8 billion in the future. 00:09:43.710 --> 00:09:46.940 align:middle line:84% So now what we are going to do is actually thinking about all 00:09:46.940 --> 00:09:51.110 align:middle line:84% of these framework to assess whether it is worth 00:09:51.110 --> 00:09:53.930 align:middle line:84% investing in preventing those damages or not, 00:09:53.930 --> 00:09:56.280 align:middle line:84% given the predictions of the city of Boston. 00:09:56.280 --> 00:09:57.080 align:middle line:90% [INAUDIBLE]? 00:09:57.080 --> 00:10:00.920 align:middle line:84% AUDIENCE: We saw a similar table yesterday in studio from-- 00:10:00.920 --> 00:10:02.210 align:middle line:90% a city official came in. 00:10:02.210 --> 00:10:03.390 align:middle line:90% I asked a similar question. 00:10:03.390 --> 00:10:05.280 align:middle line:84% He wasn't able to give me a direct answer. 00:10:05.280 --> 00:10:08.840 align:middle line:84% So this, the same study from 10 years ago, 00:10:08.840 --> 00:10:11.180 align:middle line:84% predicted 9 inches of sea level rise 00:10:11.180 --> 00:10:13.370 align:middle line:84% by 2030, which is only 7 years from now. 00:10:13.370 --> 00:10:15.450 align:middle line:90% So we're past the halfway point. 00:10:15.450 --> 00:10:18.380 align:middle line:84% Have we seen 4.5 inches of sea level rise? 00:10:18.380 --> 00:10:21.110 align:middle line:84% Is there actually $1.2 billion in damages 00:10:21.110 --> 00:10:23.703 align:middle line:84% that have been experienced or 50% of that, $600 million 00:10:23.703 --> 00:10:25.620 align:middle line:84% in damages that have been experienced to date? 00:10:25.620 --> 00:10:28.460 align:middle line:84% Or is this being updated with new projections 00:10:28.460 --> 00:10:31.170 align:middle line:84% to reflect what's happened over the last 10 years? 00:10:31.170 --> 00:10:35.060 align:middle line:84% JUAN PALACIOS: Every report that you have there of this kind 00:10:35.060 --> 00:10:37.470 align:middle line:84% will be always using the latest prediction. 00:10:37.470 --> 00:10:40.070 align:middle line:84% And you saw the table of the beliefs 00:10:40.070 --> 00:10:43.430 align:middle line:84% of the IPCC, the United Nations regarding climate change. 00:10:43.430 --> 00:10:46.950 align:middle line:84% That is updated every four or five years. 00:10:46.950 --> 00:10:50.780 align:middle line:84% So especially in 2022, there was a big update 00:10:50.780 --> 00:10:52.700 align:middle line:84% around those projections of the damages. 00:10:52.700 --> 00:10:57.320 align:middle line:84% Here, what you will see is these are the-- usually, what you're 00:10:57.320 --> 00:10:59.960 align:middle line:84% getting to get these numbers is actually the properties that 00:10:59.960 --> 00:11:02.252 align:middle line:84% are going to be exposed or the properties that are here 00:11:02.252 --> 00:11:05.840 align:middle line:84% in this blue area times the value of those properties, 00:11:05.840 --> 00:11:07.875 align:middle line:84% and that's how they calculate the $8 billion. 00:11:07.875 --> 00:11:10.220 align:middle line:84% AUDIENCE: Right, but the study itself, though, 00:11:10.220 --> 00:11:11.697 align:middle line:90% is done-- this one dated 2013. 00:11:11.697 --> 00:11:12.780 align:middle line:90% The one we saw yesterday-- 00:11:12.780 --> 00:11:13.740 align:middle line:90% 2017. 00:11:13.740 --> 00:11:15.950 align:middle line:84% Both showed 9 inches of relative sea rise 00:11:15.950 --> 00:11:19.160 align:middle line:84% by 2030, which is 7 years from now. 00:11:19.160 --> 00:11:22.590 align:middle line:84% Have we experienced more than 50% of that increase to date? 00:11:22.590 --> 00:11:25.555 align:middle line:84% I was wondering what the prediction was 00:11:25.555 --> 00:11:27.680 align:middle line:84% 10 years ago-- has that actually come into fruition 00:11:27.680 --> 00:11:31.550 align:middle line:84% or is there a need for more conservative estimates 00:11:31.550 --> 00:11:34.680 align:middle line:84% as to what the damages actually will be by that year? 00:11:34.680 --> 00:11:37.040 align:middle line:84% JUAN PALACIOS: Yeah, there are two things there. 00:11:37.040 --> 00:11:40.760 align:middle line:84% I don't know exactly the numbers of how much of these 9 inches 00:11:40.760 --> 00:11:42.210 align:middle line:90% have been already observed. 00:11:42.210 --> 00:11:44.900 align:middle line:84% There is some sea level rise that has been, of course, 00:11:44.900 --> 00:11:46.050 align:middle line:90% observed there. 00:11:46.050 --> 00:11:49.750 align:middle line:84% There is a risk of doing that type of logic. 00:11:49.750 --> 00:11:54.020 align:middle line:84% That is that you assume some of linear increase in sea level 00:11:54.020 --> 00:11:54.520 align:middle line:90% rise. 00:11:54.520 --> 00:11:57.070 align:middle line:84% Every year, I get more and more and more and more, 00:11:57.070 --> 00:11:59.780 align:middle line:84% and the same delta will apply every year. 00:11:59.780 --> 00:12:01.240 align:middle line:84% But actually, a lot of these things 00:12:01.240 --> 00:12:03.490 align:middle line:84% are not linear because if you get-- 00:12:03.490 --> 00:12:08.200 align:middle line:84% if you try to visualize how things in Greenland, 00:12:08.200 --> 00:12:11.440 align:middle line:84% for example, work when it comes to melting of ice, 00:12:11.440 --> 00:12:14.350 align:middle line:84% these are big chunks of ice that eventually 00:12:14.350 --> 00:12:16.820 align:middle line:90% will come into the seawater. 00:12:16.820 --> 00:12:19.833 align:middle line:84% And those doesn't happen kind of the same chunk of ice 00:12:19.833 --> 00:12:21.250 align:middle line:84% happens-- actually, the worst that 00:12:21.250 --> 00:12:22.792 align:middle line:84% can happen in terms of sea level rise 00:12:22.792 --> 00:12:27.760 align:middle line:84% that there is this big parts of ice of places like Greenland 00:12:27.760 --> 00:12:30.310 align:middle line:84% that become-- all of the sudden detach 00:12:30.310 --> 00:12:32.030 align:middle line:90% from the big block of ice. 00:12:32.030 --> 00:12:35.200 align:middle line:84% So it can be that it is, like say here, 00:12:35.200 --> 00:12:38.860 align:middle line:84% you have 2 inch or 3 inches happening for the period of 5-6 00:12:38.860 --> 00:12:41.740 align:middle line:84% years, and all of the sudden, there is such an event that they 00:12:41.740 --> 00:12:45.280 align:middle line:84% call tipping points, and then you go up to 7 inches. 00:12:45.280 --> 00:12:49.780 align:middle line:84% So I think that it is hard to back up 00:12:49.780 --> 00:12:52.510 align:middle line:84% a lot of these issues with empirical observations 00:12:52.510 --> 00:12:55.450 align:middle line:84% on a year-to-year basis because a lot of these observations 00:12:55.450 --> 00:12:57.730 align:middle line:84% happen via very thorough assessments 00:12:57.730 --> 00:13:02.080 align:middle line:84% of de-icing and different events that you know 00:13:02.080 --> 00:13:04.900 align:middle line:84% will end up happening, and that will end up 00:13:04.900 --> 00:13:07.180 align:middle line:84% having a consequence on the sea level rise. 00:13:07.180 --> 00:13:08.700 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:13:08.700 --> 00:13:09.960 align:middle line:90% JUAN PALACIOS: Yeah? 00:13:09.960 --> 00:13:12.660 align:middle line:84% AUDIENCE: Just, like, for background [INAUDIBLE] 00:13:12.660 --> 00:13:17.640 align:middle line:84% it has actually risen 8 inches over the past 10 years. 00:13:17.640 --> 00:13:19.860 align:middle line:84% And there's some data I find online 00:13:19.860 --> 00:13:22.690 align:middle line:84% that shows the actual numbers year by year. 00:13:22.690 --> 00:13:27.778 align:middle line:84% So yeah, I think it's based on historical data and projected 00:13:27.778 --> 00:13:28.882 align:middle line:90% data. 00:13:28.882 --> 00:13:29.840 align:middle line:90% JUAN PALACIOS: Awesome. 00:13:29.840 --> 00:13:32.660 align:middle line:84% And here is the plan of the city to invest 00:13:32.660 --> 00:13:36.530 align:middle line:84% in a lot of protection mechanisms to avoid that these-- 00:13:36.530 --> 00:13:37.470 align:middle line:90% you see low battery. 00:13:37.470 --> 00:13:39.860 align:middle line:84% Everyone has low battery, including myself. 00:13:39.860 --> 00:13:42.560 align:middle line:90% So here it is, the-- 00:13:42.560 --> 00:13:45.950 align:middle line:84% let me see what I can get rid of it. 00:13:45.950 --> 00:13:46.640 align:middle line:90% No worries. 00:13:46.640 --> 00:13:49.280 align:middle line:84% It's low battery, but not that low. 00:13:49.280 --> 00:13:53.690 align:middle line:84% These are the costs associated with protecting sea port. 00:13:53.690 --> 00:13:55.460 align:middle line:90% You have Fort Point Channel. 00:13:55.460 --> 00:13:58.970 align:middle line:84% You have different waterfront, different city walls 00:13:58.970 --> 00:14:00.800 align:middle line:84% that you will put all over the area 00:14:00.800 --> 00:14:03.920 align:middle line:84% to make sure that there is no such thing as those damages 00:14:03.920 --> 00:14:05.870 align:middle line:84% that we were projected in the previous one 00:14:05.870 --> 00:14:08.000 align:middle line:84% to avoid that the properties and the people 00:14:08.000 --> 00:14:09.200 align:middle line:90% are exposed to that one. 00:14:09.200 --> 00:14:11.490 align:middle line:84% And you see that it's quite a lot of money here. 00:14:11.490 --> 00:14:17.330 align:middle line:84% You have a total of around from $500 million to $1 billion, 00:14:17.330 --> 00:14:21.560 align:middle line:84% and going to $7 million to $15 million in maintenance 00:14:21.560 --> 00:14:23.810 align:middle line:84% because once you build that infrastructure, of course, 00:14:23.810 --> 00:14:25.422 align:middle line:90% you cannot stay there forever. 00:14:25.422 --> 00:14:26.880 align:middle line:84% You need to maintain it every year. 00:14:26.880 --> 00:14:30.570 align:middle line:84% So it's a significant investment that you have. 00:14:30.570 --> 00:14:34.390 align:middle line:84% And $1 billion, let's be pessimistic and conservative, 00:14:34.390 --> 00:14:36.000 align:middle line:84% you will have to invest $1 billion 00:14:36.000 --> 00:14:40.470 align:middle line:84% to protect the city against $8 billion expected damages. 00:14:40.470 --> 00:14:43.260 align:middle line:84% So when what you see is that on damages 00:14:43.260 --> 00:14:45.410 align:middle line:90% that will happen all the way. 00:14:45.410 --> 00:14:47.400 align:middle line:90% to 2070. 00:14:47.400 --> 00:14:49.550 align:middle line:84% So again, what discount rate you will 00:14:49.550 --> 00:14:52.100 align:middle line:84% put will make this investment worth it 00:14:52.100 --> 00:14:54.840 align:middle line:84% or not because you need to start investing those things now. 00:14:54.840 --> 00:14:57.980 align:middle line:84% And these damages will only happen severely 00:14:57.980 --> 00:14:59.520 align:middle line:90% over the next decades. 00:14:59.520 --> 00:15:00.830 align:middle line:90% So yeah? 00:15:00.830 --> 00:15:04.032 align:middle line:84% AUDIENCE: Are the damages [INAUDIBLE] present value? 00:15:04.032 --> 00:15:06.115 align:middle line:84% JUAN PALACIOS: No, they are when they will happen. 00:15:06.115 --> 00:15:08.750 align:middle line:90% 00:15:08.750 --> 00:15:13.940 align:middle line:84% So here is what comes down to-- so if you think about investing 00:15:13.940 --> 00:15:19.430 align:middle line:84% $1 billion to protect the city for $8 billion of damages 00:15:19.430 --> 00:15:22.640 align:middle line:84% in 50 years, you see that any changes in the cash flow 00:15:22.640 --> 00:15:25.010 align:middle line:84% will actually make a big difference 00:15:25.010 --> 00:15:26.390 align:middle line:90% into the decision-making. 00:15:26.390 --> 00:15:29.700 align:middle line:84% We have always, in this course, NPV positive and NPV negative. 00:15:29.700 --> 00:15:35.000 align:middle line:84% And anything that you have here from 7% to 1%, 00:15:35.000 --> 00:15:37.940 align:middle line:84% that is a lot of the discount rates that you end up 00:15:37.940 --> 00:15:39.710 align:middle line:90% seeing in the literature. 00:15:39.710 --> 00:15:42.150 align:middle line:84% You see that there will be a very different outcome. 00:15:42.150 --> 00:15:45.530 align:middle line:84% So anything that will be below 4 will actually 00:15:45.530 --> 00:15:49.100 align:middle line:84% make it that those damages will be counting enough at present 00:15:49.100 --> 00:15:51.740 align:middle line:84% and will actually make it big enough 00:15:51.740 --> 00:15:54.080 align:middle line:84% to make sure that this number is above $1 million 00:15:54.080 --> 00:15:56.820 align:middle line:84% and, therefore, make an NPV positive solution. 00:15:56.820 --> 00:15:59.420 align:middle line:84% And that's where a lot of the discussion and a lot 00:15:59.420 --> 00:16:04.220 align:middle line:84% of the decisions need to be made when it comes to mitigation, 00:16:04.220 --> 00:16:05.670 align:middle line:90% resiliency, and so forth. 00:16:05.670 --> 00:16:09.920 align:middle line:84% This is not mitigation in terms of reducing CO2 emissions 00:16:09.920 --> 00:16:13.470 align:middle line:84% because that will be even an extra loop of complexity. 00:16:13.470 --> 00:16:15.500 align:middle line:84% But already just thinking about resiliency 00:16:15.500 --> 00:16:18.350 align:middle line:84% versus expected losses and getting hit, 00:16:18.350 --> 00:16:20.030 align:middle line:84% you see that the political decision 00:16:20.030 --> 00:16:22.850 align:middle line:84% is how much do we value those or how much do we believe 00:16:22.850 --> 00:16:24.600 align:middle line:90% in those things to happen. 00:16:24.600 --> 00:16:26.390 align:middle line:84% And then, ultimately, even if we have 00:16:26.390 --> 00:16:29.480 align:middle line:84% an assessment of the damages and we all agree in the room 00:16:29.480 --> 00:16:32.150 align:middle line:84% that those damages are there, then the next question 00:16:32.150 --> 00:16:35.060 align:middle line:84% is, how do we discount those to the present to make sure 00:16:35.060 --> 00:16:37.380 align:middle line:90% that we have a fair evaluation? 00:16:37.380 --> 00:16:43.580 align:middle line:84% If you see the actual data, you have these papers now evaluating 00:16:43.580 --> 00:16:45.120 align:middle line:90% these very long discount rates. 00:16:45.120 --> 00:16:46.770 align:middle line:84% It's a very complex exercise, actually. 00:16:46.770 --> 00:16:49.640 align:middle line:84% They do it with housing market because you can actually 00:16:49.640 --> 00:16:52.220 align:middle line:84% use the differences in lease terms 00:16:52.220 --> 00:16:55.670 align:middle line:84% to actually see how people will value those cash flows 00:16:55.670 --> 00:16:57.120 align:middle line:90% in the very, very far future. 00:16:57.120 --> 00:17:00.680 align:middle line:84% So, for example, I think Singapore has leases of 90-100 00:17:00.680 --> 00:17:01.350 align:middle line:90% years. 00:17:01.350 --> 00:17:04.010 align:middle line:84% You can compare those with perpetuity type of things to see 00:17:04.010 --> 00:17:08.060 align:middle line:84% how people will value the 100-year losses versus perpetual 00:17:08.060 --> 00:17:08.599 align:middle line:90% losses. 00:17:08.599 --> 00:17:11.280 align:middle line:84% And you see that the discount rate that is usually there 00:17:11.280 --> 00:17:16.230 align:middle line:90% is 2.6 in the literature. 00:17:16.230 --> 00:17:16.986 align:middle line:90% [INAUDIBLE]? 00:17:16.986 --> 00:17:18.569 align:middle line:84% AUDIENCE: This discount rate is trying 00:17:18.569 --> 00:17:23.040 align:middle line:84% to capture how the end users of the property 00:17:23.040 --> 00:17:25.444 align:middle line:84% would value this because it's the government itself 00:17:25.444 --> 00:17:26.569 align:middle line:90% that's spending this money? 00:17:26.569 --> 00:17:27.060 align:middle line:90% JUAN PALACIOS: Correct? 00:17:27.060 --> 00:17:28.685 align:middle line:84% AUDIENCE: The government doesn't really 00:17:28.685 --> 00:17:30.090 align:middle line:84% require positive NPV [INAUDIBLE]. 00:17:30.090 --> 00:17:31.540 align:middle line:90% JUAN PALACIOS: Exactly. 00:17:31.540 --> 00:17:32.040 align:middle line:90% Exactly. 00:17:32.040 --> 00:17:35.580 align:middle line:84% But you want to the ultimate homeowners-- 00:17:35.580 --> 00:17:36.690 align:middle line:90% what will they value? 00:17:36.690 --> 00:17:40.490 align:middle line:84% Because even though you, as a government, one of your missions 00:17:40.490 --> 00:17:43.460 align:middle line:84% is to do NPV-negative investments there 00:17:43.460 --> 00:17:45.300 align:middle line:84% because, otherwise, it wouldn't happen. 00:17:45.300 --> 00:17:48.110 align:middle line:84% You want to have some alignment of what these people will value 00:17:48.110 --> 00:17:52.100 align:middle line:84% and so forth to avoid burning your political power. 00:17:52.100 --> 00:17:52.860 align:middle line:90% Julio. 00:17:52.860 --> 00:17:55.152 align:middle line:84% AUDIENCE: I'm sorry, I'm just trying to get this right. 00:17:55.152 --> 00:17:59.280 align:middle line:84% So the $8 billion is the damages of the properties 00:17:59.280 --> 00:18:00.498 align:middle line:90% that are there right now. 00:18:00.498 --> 00:18:01.290 align:middle line:90% JUAN PALACIOS: Yes. 00:18:01.290 --> 00:18:04.530 align:middle line:84% AUDIENCE: But you could have more properties or taller 00:18:04.530 --> 00:18:06.870 align:middle line:84% buildings, so the building could actually be a lot more? 00:18:06.870 --> 00:18:08.787 align:middle line:84% JUAN PALACIOS: In seaport I think not anymore. 00:18:08.787 --> 00:18:11.430 align:middle line:84% I think that everything is fairly developed already. 00:18:11.430 --> 00:18:12.724 align:middle line:90% This these for seaport. 00:18:12.724 --> 00:18:13.516 align:middle line:90% AUDIENCE: Then 50-- 00:18:13.516 --> 00:18:16.205 align:middle line:84% SIQI ZHENG: [INAUDIBLE] vacant land. 00:18:16.205 --> 00:18:17.580 align:middle line:84% JUAN PALACIOS: OK, so there might 00:18:17.580 --> 00:18:20.252 align:middle line:84% be a so there might be that even there might be some more 00:18:20.252 --> 00:18:21.960 align:middle line:84% damages because you develop more property 00:18:21.960 --> 00:18:25.210 align:middle line:84% and you put more capital there, yeah. 00:18:25.210 --> 00:18:27.640 align:middle line:84% So that will make it even worse because it will not be $8, 00:18:27.640 --> 00:18:30.182 align:middle line:84% but it will be $10 if you keep developing high rises in areas 00:18:30.182 --> 00:18:31.610 align:middle line:90% that will be eventually hit by-- 00:18:31.610 --> 00:18:33.110 align:middle line:84% AUDIENCE: [INAUDIBLE] more feasible. 00:18:33.110 --> 00:18:35.830 align:middle line:84% JUAN PALACIOS: Yeah, exactly because then the $8 00:18:35.830 --> 00:18:37.960 align:middle line:84% will become $10, so then everything will be-- 00:18:37.960 --> 00:18:41.080 align:middle line:84% and this is like a public good for the entire district. 00:18:41.080 --> 00:18:43.930 align:middle line:84% Doesn't matter whether there are, here, 5, 10, 20, 00:18:43.930 --> 00:18:45.100 align:middle line:90% 300 buildings. 00:18:45.100 --> 00:18:47.470 align:middle line:84% You are trying to protect the whole border 00:18:47.470 --> 00:18:49.370 align:middle line:90% and protect everyone inside. 00:18:49.370 --> 00:18:53.650 align:middle line:84% So actually, the emptier it is, the more difficult 00:18:53.650 --> 00:18:55.870 align:middle line:84% it's going to justify the investment because the less 00:18:55.870 --> 00:18:57.525 align:middle line:90% damages are going to be there. 00:18:57.525 --> 00:18:58.900 align:middle line:84% And now we're going to keep going 00:18:58.900 --> 00:19:01.390 align:middle line:84% with all of these discounts and so forth with [INAUDIBLE] 00:19:01.390 --> 00:19:03.280 align:middle line:84% and how we are actually going to think 00:19:03.280 --> 00:19:07.180 align:middle line:84% about those beliefs in a more thorough way. 00:19:07.180 --> 00:19:08.303 align:middle line:90% Exactly 10 minutes. 00:19:08.303 --> 00:19:09.220 align:middle line:90% SIQI ZHENG: Yes, good. 00:19:09.220 --> 00:19:10.300 align:middle line:90% We switch. 00:19:10.300 --> 00:19:12.730 align:middle line:84% I could [INAUDIBLE] the last question 00:19:12.730 --> 00:19:17.240 align:middle line:84% I think [INAUDIBLE] very interesting [INAUDIBLE] 00:19:17.240 --> 00:19:22.790 align:middle line:84% protection [INAUDIBLE] sea wall and all the things. 00:19:22.790 --> 00:19:25.250 align:middle line:84% And if the government invests a lot, 00:19:25.250 --> 00:19:27.740 align:middle line:84% then that will boost the confidence 00:19:27.740 --> 00:19:29.330 align:middle line:90% of the private sector. 00:19:29.330 --> 00:19:32.728 align:middle line:84% The private sector will say, oh, this is not a big problem. 00:19:32.728 --> 00:19:33.770 align:middle line:90% You can just [INAUDIBLE]. 00:19:33.770 --> 00:19:35.840 align:middle line:90% You can further develop. 00:19:35.840 --> 00:19:37.700 align:middle line:90% So now it's a debate. 00:19:37.700 --> 00:19:39.360 align:middle line:90% It's an endogenous thing. 00:19:39.360 --> 00:19:41.120 align:middle line:84% If the government invests a lot of money 00:19:41.120 --> 00:19:44.120 align:middle line:84% to protect that area, then the developers, 00:19:44.120 --> 00:19:45.950 align:middle line:84% the individual investors, they feel 00:19:45.950 --> 00:19:49.220 align:middle line:84% more confident to further invest in that area, 00:19:49.220 --> 00:19:52.130 align:middle line:90% to build buildings like that. 00:19:52.130 --> 00:19:55.610 align:middle line:84% So that may lead to a misallocation of resources 00:19:55.610 --> 00:19:57.410 align:middle line:84% because the cities, the city mayors, 00:19:57.410 --> 00:19:59.640 align:middle line:84% they always want to protect their cities. 00:19:59.640 --> 00:20:02.900 align:middle line:84% And then those dangerous places, risky places, 00:20:02.900 --> 00:20:09.390 align:middle line:84% they got more and more private investors to build there. 00:20:09.390 --> 00:20:11.640 align:middle line:90% Anyway, now, let me continue. 00:20:11.640 --> 00:20:17.870 align:middle line:84% So, basically, I think-- we are the first real estate program. 00:20:17.870 --> 00:20:20.420 align:middle line:84% I'm very proud of this, that we have this sustainable real 00:20:20.420 --> 00:20:22.560 align:middle line:90% estate as a core class. 00:20:22.560 --> 00:20:23.660 align:middle line:90% Maybe you like it. 00:20:23.660 --> 00:20:24.660 align:middle line:90% Maybe you don't like it. 00:20:24.660 --> 00:20:27.980 align:middle line:84% But we are the first real estate master's program 00:20:27.980 --> 00:20:30.380 align:middle line:84% that we have this sustainable real estate as a core. 00:20:30.380 --> 00:20:32.510 align:middle line:84% And also, I think we are the first 00:20:32.510 --> 00:20:37.160 align:middle line:84% that we have three sessions dedicated to climate 00:20:37.160 --> 00:20:39.630 align:middle line:84% because we know climate is so important. 00:20:39.630 --> 00:20:42.360 align:middle line:84% And last year, we had two and we couldn't finish. 00:20:42.360 --> 00:20:44.010 align:middle line:90% So this year, we have three. 00:20:44.010 --> 00:20:46.160 align:middle line:90% So we are not in a hurry. 00:20:46.160 --> 00:20:47.040 align:middle line:90% We're not in a hurry. 00:20:47.040 --> 00:20:50.420 align:middle line:84% We are going to enjoy the process of these three lectures 00:20:50.420 --> 00:20:51.480 align:middle line:90% all about climate. 00:20:51.480 --> 00:20:53.930 align:middle line:84% So, after the three sessions, you 00:20:53.930 --> 00:20:56.060 align:middle line:84% become expert on climate real estate. 00:20:56.060 --> 00:20:57.440 align:middle line:90% So that's the point. 00:20:57.440 --> 00:20:59.550 align:middle line:90% So, today, I will continue. 00:20:59.550 --> 00:21:03.050 align:middle line:84% And then, next week, we will have another last session, 00:21:03.050 --> 00:21:06.080 align:middle line:84% and I'm going to invite our guest speaker. 00:21:06.080 --> 00:21:08.270 align:middle line:84% I don't want to stand here always talking-- 00:21:08.270 --> 00:21:10.320 align:middle line:90% too boring. 00:21:10.320 --> 00:21:16.460 align:middle line:84% Last week, last session, we have Moody's cabin. 00:21:16.460 --> 00:21:20.920 align:middle line:84% Then, next Tuesday, we will have a very important person 00:21:20.920 --> 00:21:26.370 align:middle line:84% who is a former head of the FHA, Federal Housing Authority. 00:21:26.370 --> 00:21:31.900 align:middle line:84% So that's a big federal housing administration, and also 00:21:31.900 --> 00:21:34.510 align:middle line:84% the senior leadership in the Fannie Mae and Freddie 00:21:34.510 --> 00:21:37.880 align:middle line:84% Mac for many, many years at [INAUDIBLE]. 00:21:37.880 --> 00:21:39.380 align:middle line:90% And he will come here. 00:21:39.380 --> 00:21:41.590 align:middle line:84% He is a senior lecturer at the Sloan 00:21:41.590 --> 00:21:46.220 align:middle line:84% School heading some housing research center there. 00:21:46.220 --> 00:21:50.350 align:middle line:84% And he will talk give a guest talk next Tuesday 00:21:50.350 --> 00:21:52.660 align:middle line:84% on climate and real estate and how, 00:21:52.660 --> 00:21:58.180 align:middle line:84% from this government perspective, this HUD and FHA 00:21:58.180 --> 00:22:00.730 align:middle line:84% and Fannie Mae, how they are thinking of this 00:22:00.730 --> 00:22:03.760 align:middle line:84% and whether they want to invest more money, public money, 00:22:03.760 --> 00:22:06.400 align:middle line:84% to help those places that got flooded. 00:22:06.400 --> 00:22:07.870 align:middle line:90% So that will be next Tuesday. 00:22:07.870 --> 00:22:09.980 align:middle line:90% But today, will be Juan and me. 00:22:09.980 --> 00:22:11.150 align:middle line:90% Now it's my turn. 00:22:11.150 --> 00:22:16.290 align:middle line:84% So this is the second lecture on climate and real estate. 00:22:16.290 --> 00:22:18.720 align:middle line:84% So just now, I think I can continue just 00:22:18.720 --> 00:22:21.720 align:middle line:84% now the conversation because just now, 00:22:21.720 --> 00:22:24.180 align:middle line:84% we talk about discount climate risk, 00:22:24.180 --> 00:22:28.290 align:middle line:84% and then we know that climate risk is a future risk. 00:22:28.290 --> 00:22:29.740 align:middle line:90% So it's not today. 00:22:29.740 --> 00:22:30.520 align:middle line:90% It's not today. 00:22:30.520 --> 00:22:37.290 align:middle line:84% It's about future-- next 50 years, next 100 years. 00:22:37.290 --> 00:22:38.468 align:middle line:90% So where's the pointer? 00:22:38.468 --> 00:22:39.760 align:middle line:90% JUAN PALACIOS: It's right here. 00:22:39.760 --> 00:22:42.524 align:middle line:90% 00:22:42.524 --> 00:22:43.400 align:middle line:90% SIQI ZHENG: Oh, here. 00:22:43.400 --> 00:22:45.910 align:middle line:90% 00:22:45.910 --> 00:22:50.040 align:middle line:84% Because the real estate is a long-term thing-- 00:22:50.040 --> 00:22:51.620 align:middle line:90% it's not a one day-- 00:22:51.620 --> 00:22:54.800 align:middle line:84% a long-term thing, holding period, many, many years. 00:22:54.800 --> 00:23:00.300 align:middle line:84% So we all understand the analysis is about future. 00:23:00.300 --> 00:23:04.460 align:middle line:84% When we talk about climate risk, always about future things. 00:23:04.460 --> 00:23:07.130 align:middle line:84% That is this discount rate just now 00:23:07.130 --> 00:23:09.000 align:middle line:84% Juan mentioned that is so important. 00:23:09.000 --> 00:23:13.430 align:middle line:84% If you choose 2.6% or you choose 3% or you 00:23:13.430 --> 00:23:19.610 align:middle line:84% choose 5%, totally different and will yield different NPVs. 00:23:19.610 --> 00:23:23.630 align:middle line:84% So that's the point of this discount rate-- so important. 00:23:23.630 --> 00:23:29.900 align:middle line:84% So that's this IPCC, that's RCP, Representative Concentration 00:23:29.900 --> 00:23:30.560 align:middle line:90% Pathway. 00:23:30.560 --> 00:23:34.890 align:middle line:84% And RCP 8.5 is very serious global warming. 00:23:34.890 --> 00:23:38.810 align:middle line:84% And this one is, like, if the human beings can mitigate. 00:23:38.810 --> 00:23:42.590 align:middle line:84% So then think about this hypothetically. 00:23:42.590 --> 00:23:45.740 align:middle line:90% You buy an asset 2020. 00:23:45.740 --> 00:23:50.340 align:middle line:84% Then, maybe you will just hold the asset for many, many years-- 00:23:50.340 --> 00:23:51.240 align:middle line:90% you sell. 00:23:51.240 --> 00:23:57.570 align:middle line:84% So now when you are evaluating the impact of climate risk, 00:23:57.570 --> 00:23:59.910 align:middle line:90% you are thinking like this way. 00:23:59.910 --> 00:24:02.160 align:middle line:84% You are thinking because the global warming 00:24:02.160 --> 00:24:03.780 align:middle line:90% becomes more and more serious-- 00:24:03.780 --> 00:24:05.410 align:middle line:90% now, maybe no problem. 00:24:05.410 --> 00:24:07.890 align:middle line:84% Just not when it shows the seaport. 00:24:07.890 --> 00:24:09.610 align:middle line:90% For now, no problem. 00:24:09.610 --> 00:24:12.000 align:middle line:84% But then, little by little, more and more and more 00:24:12.000 --> 00:24:15.120 align:middle line:84% hurricanes and the sea level rise and all kinds of things 00:24:15.120 --> 00:24:15.960 align:middle line:90% happen. 00:24:15.960 --> 00:24:25.590 align:middle line:84% So conceptually, we say, OK, when me, as a buyer in 2020, 00:24:25.590 --> 00:24:30.210 align:middle line:84% I stand here at this moment to consider whether to buy or not 00:24:30.210 --> 00:24:31.770 align:middle line:90% and my bidding price. 00:24:31.770 --> 00:24:33.070 align:middle line:90% So I'm standing here. 00:24:33.070 --> 00:24:37.740 align:middle line:84% So my current bid for asset is the present value 00:24:37.740 --> 00:24:39.450 align:middle line:90% of the future cash flows-- 00:24:39.450 --> 00:24:42.270 align:middle line:84% and, for example, all other things-- 00:24:42.270 --> 00:24:46.530 align:middle line:84% and minus the present value of the expected climate damages. 00:24:46.530 --> 00:24:47.760 align:middle line:90% Who knows? 00:24:47.760 --> 00:24:50.340 align:middle line:84% All the scientists-- they publish so many papers, 00:24:50.340 --> 00:24:51.750 align:middle line:90% but they are all guessing. 00:24:51.750 --> 00:24:54.100 align:middle line:84% They're making all kinds of projections. 00:24:54.100 --> 00:24:58.650 align:middle line:84% So this expected loss so important. 00:24:58.650 --> 00:25:02.550 align:middle line:84% And then you say, OK, from the scientific report, 00:25:02.550 --> 00:25:05.790 align:middle line:84% we also wrote many, many papers and we published, and so 00:25:05.790 --> 00:25:09.910 align:middle line:84% many climate scientists, they say, oh, cost of damages-- 00:25:09.910 --> 00:25:11.970 align:middle line:84% they say this-- every year, how many 00:25:11.970 --> 00:25:14.070 align:middle line:90% the likelihood of the hurricane? 00:25:14.070 --> 00:25:17.290 align:middle line:90% Then the thing is, you can-- 00:25:17.290 --> 00:25:19.020 align:middle line:90% two factors here. 00:25:19.020 --> 00:25:20.790 align:middle line:84% Discount rate just now we mentioned-- 00:25:20.790 --> 00:25:22.050 align:middle line:90% this is so important. 00:25:22.050 --> 00:25:24.460 align:middle line:90% 2%, 3%-- totally different. 00:25:24.460 --> 00:25:26.220 align:middle line:90% So Juan covered this. 00:25:26.220 --> 00:25:28.650 align:middle line:90% Now we are looking at this. 00:25:28.650 --> 00:25:32.910 align:middle line:84% Scientists are saying so, but not all people will believe in. 00:25:32.910 --> 00:25:34.960 align:middle line:90% You know it's a climate belief. 00:25:34.960 --> 00:25:38.573 align:middle line:84% So when you buy the asset and we think about future, 00:25:38.573 --> 00:25:39.990 align:middle line:84% you are thinking whether later you 00:25:39.990 --> 00:25:42.660 align:middle line:84% can sell that to the second buyer, the later one who 00:25:42.660 --> 00:25:43.750 align:middle line:90% will buy this from you. 00:25:43.750 --> 00:25:49.170 align:middle line:84% So all this market, who knows, even if the scientist publish 00:25:49.170 --> 00:25:51.060 align:middle line:90% all these reports? 00:25:51.060 --> 00:25:54.420 align:middle line:84% And maybe half of the people, the buyers-- 00:25:54.420 --> 00:25:55.780 align:middle line:90% they believe in that. 00:25:55.780 --> 00:25:57.490 align:middle line:84% The other half-- they don't believe. 00:25:57.490 --> 00:26:01.570 align:middle line:84% And this one will definitely influence the buyers, 00:26:01.570 --> 00:26:04.900 align:middle line:84% future buyers, the market participants, 00:26:04.900 --> 00:26:08.000 align:middle line:84% which means we cannot just look at this numbers. 00:26:08.000 --> 00:26:10.760 align:middle line:90% We must look at the belief. 00:26:10.760 --> 00:26:14.630 align:middle line:84% So, fortunately, we also have some scientific measures 00:26:14.630 --> 00:26:15.660 align:middle line:90% of belief. 00:26:15.660 --> 00:26:18.740 align:middle line:84% So there's a very famous saying-- you can Google this. 00:26:18.740 --> 00:26:23.940 align:middle line:84% And Yale University, they have a Yale Climate Opinion map. 00:26:23.940 --> 00:26:26.040 align:middle line:90% They have a research group. 00:26:26.040 --> 00:26:28.340 align:middle line:84% They already run for many, many years. 00:26:28.340 --> 00:26:32.130 align:middle line:84% And each year-- they have a lot of research funding-- 00:26:32.130 --> 00:26:36.680 align:middle line:84% I think they just did a huge large-scale survey, thousands 00:26:36.680 --> 00:26:38.490 align:middle line:90% of hundreds of US citizens. 00:26:38.490 --> 00:26:42.260 align:middle line:84% I think now they are only looking at the United States. 00:26:42.260 --> 00:26:44.670 align:middle line:90% So that's basically this. 00:26:44.670 --> 00:26:46.680 align:middle line:84% And then you go to the county level. 00:26:46.680 --> 00:26:48.890 align:middle line:84% Now this is a county-level thing. 00:26:48.890 --> 00:26:51.520 align:middle line:84% You see the county and the metropolitan area 00:26:51.520 --> 00:26:53.000 align:middle line:90% stays aggregate. 00:26:53.000 --> 00:26:54.080 align:middle line:90% Then they go there. 00:26:54.080 --> 00:26:57.360 align:middle line:84% They interview, and they have a bunch of questions. 00:26:57.360 --> 00:26:59.220 align:middle line:90% And then you have this map. 00:26:59.220 --> 00:27:03.560 align:middle line:84% They provide indexes, and then we can always use these indexes 00:27:03.560 --> 00:27:05.010 align:middle line:90% for our own research. 00:27:05.010 --> 00:27:07.488 align:middle line:84% So then you say, OK, this is 100%. 00:27:07.488 --> 00:27:09.530 align:middle line:84% So the question-- they have many, many questions. 00:27:09.530 --> 00:27:10.830 align:middle line:90% This is one of the questions. 00:27:10.830 --> 00:27:14.380 align:middle line:84% For adults who think global warming is happening. 00:27:14.380 --> 00:27:18.340 align:middle line:84% Then for the thing is, for the entire United States, 00:27:18.340 --> 00:27:22.790 align:middle line:84% 72% of the adults of us-- we are adults-- 00:27:22.790 --> 00:27:25.720 align:middle line:84% we think global warming is happening. 00:27:25.720 --> 00:27:29.890 align:middle line:84% And the 14%, they don't think global warming is happening. 00:27:29.890 --> 00:27:32.620 align:middle line:84% And in the middle, they don't know. 00:27:32.620 --> 00:27:33.940 align:middle line:90% They are not very sure. 00:27:33.940 --> 00:27:36.670 align:middle line:84% So that's the entire United States. 00:27:36.670 --> 00:27:39.520 align:middle line:84% Of course, you understand why, here, we are darker. 00:27:39.520 --> 00:27:42.100 align:middle line:84% Here, we maybe have a higher share. 00:27:42.100 --> 00:27:45.590 align:middle line:84% And here, the darker and in the middle, maybe lighter. 00:27:45.590 --> 00:27:47.710 align:middle line:90% So that's a situation. 00:27:47.710 --> 00:27:53.380 align:middle line:84% Which means if you hold a scientific report, like Siqi's 00:27:53.380 --> 00:27:56.290 align:middle line:84% paper, and if you go here, go here, 00:27:56.290 --> 00:27:57.920 align:middle line:90% you get a lot of supporters. 00:27:57.920 --> 00:27:58.850 align:middle line:90% Yes, that's true. 00:27:58.850 --> 00:28:02.540 align:middle line:84% If you go here, oh, don't trust someone to fake me. 00:28:02.540 --> 00:28:03.640 align:middle line:90% It's not true. 00:28:03.640 --> 00:28:05.380 align:middle line:90% I don't believe in that. 00:28:05.380 --> 00:28:07.680 align:middle line:90% Question? 00:28:07.680 --> 00:28:09.680 align:middle line:84% AUDIENCE: I think it would be interesting to put 00:28:09.680 --> 00:28:12.860 align:middle line:84% this map in comparison to voting patterns? 00:28:12.860 --> 00:28:13.837 align:middle line:90% SIQI ZHENG: Yes. 00:28:13.837 --> 00:28:15.420 align:middle line:84% AUDIENCE: I'm sure they're correlated. 00:28:15.420 --> 00:28:15.890 align:middle line:90% SIQI ZHENG: Yes. 00:28:15.890 --> 00:28:16.920 align:middle line:84% AUDIENCE: And it goes back to what 00:28:16.920 --> 00:28:18.420 align:middle line:84% you're talking about with beliefs, 00:28:18.420 --> 00:28:20.430 align:middle line:84% and it just becomes so politicized. 00:28:20.430 --> 00:28:22.730 align:middle line:84% Even the word global warming versus climate change 00:28:22.730 --> 00:28:26.145 align:middle line:84% and how people use the wording and thinking about, OK, then 00:28:26.145 --> 00:28:28.520 align:middle line:84% how are we going to use discount rates because people who 00:28:28.520 --> 00:28:31.500 align:middle line:84% don't believe in it or don't think it's as severe 00:28:31.500 --> 00:28:33.750 align:middle line:84% are going to say, hey, we want a higher discount rate. 00:28:33.750 --> 00:28:35.630 align:middle line:84% And then it makes it really difficult 00:28:35.630 --> 00:28:39.600 align:middle line:84% to try and to try and act upon it. 00:28:39.600 --> 00:28:40.640 align:middle line:90% SIQI ZHENG: Very good. 00:28:40.640 --> 00:28:43.160 align:middle line:84% This can be a research topic, like a thesis topic, 00:28:43.160 --> 00:28:46.100 align:middle line:84% if anyone you can-- actually, two years ago, 00:28:46.100 --> 00:28:48.030 align:middle line:90% we had a PPP student-- 00:28:48.030 --> 00:28:49.670 align:middle line:90% the other program. 00:28:49.670 --> 00:28:53.400 align:middle line:84% He did a thesis exactly on this kind of topic. 00:28:53.400 --> 00:28:56.510 align:middle line:84% He tracked these political votes and correlate 00:28:56.510 --> 00:29:02.660 align:middle line:84% with this climate belief and other political opinions. 00:29:02.660 --> 00:29:04.710 align:middle line:90% OK, so that's a [INAUDIBLE]. 00:29:04.710 --> 00:29:08.410 align:middle line:84% And then this is Yale University's one. 00:29:08.410 --> 00:29:12.420 align:middle line:84% This is another thing I got from another research institute, 00:29:12.420 --> 00:29:14.640 align:middle line:90% and this is by country. 00:29:14.640 --> 00:29:16.000 align:middle line:90% This is by country. 00:29:16.000 --> 00:29:18.880 align:middle line:90% So Greece, number one-- 00:29:18.880 --> 00:29:23.340 align:middle line:84% 90% think global warming is a major challenge, 00:29:23.340 --> 00:29:26.800 align:middle line:84% and followed by South Korea, France, Spain. 00:29:26.800 --> 00:29:27.365 align:middle line:90% I don't know. 00:29:27.365 --> 00:29:31.840 align:middle line:90% 00:29:31.840 --> 00:29:34.380 align:middle line:90% So why South Korea? 00:29:34.380 --> 00:29:35.700 align:middle line:90% OK? 00:29:35.700 --> 00:29:36.990 align:middle line:90% Are you from South Korea? 00:29:36.990 --> 00:29:37.680 align:middle line:90% OK. 00:29:37.680 --> 00:29:40.260 align:middle line:84% Do you have any sense why South Korean people, 00:29:40.260 --> 00:29:44.110 align:middle line:84% they believe in climate risk so much? 00:29:44.110 --> 00:29:45.270 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:29:45.270 --> 00:29:46.620 align:middle line:90% SIQI ZHENG: That's good. 00:29:46.620 --> 00:29:48.400 align:middle line:90% You should be proud of this. 00:29:48.400 --> 00:29:51.290 align:middle line:90% 00:29:51.290 --> 00:29:55.860 align:middle line:84% AUDIENCE: That's because [INAUDIBLE] advertise 00:29:55.860 --> 00:29:58.620 align:middle line:84% about global warming, so people in Korea 00:29:58.620 --> 00:30:02.370 align:middle line:84% doesn't actually try to criticize them, 00:30:02.370 --> 00:30:05.770 align:middle line:84% but [INAUDIBLE] so I think people are trying to [INAUDIBLE] 00:30:05.770 --> 00:30:07.740 align:middle line:90% SIQI ZHENG: Yeah, good. 00:30:07.740 --> 00:30:09.780 align:middle line:84% And I think, for those small islands 00:30:09.780 --> 00:30:12.270 align:middle line:84% because they really have the threat, 00:30:12.270 --> 00:30:15.250 align:middle line:84% then they have more sensitivity to climate. 00:30:15.250 --> 00:30:18.720 align:middle line:84% And also related to this education and public media 00:30:18.720 --> 00:30:21.460 align:middle line:84% and all this, and also politicians' agenda. 00:30:21.460 --> 00:30:23.840 align:middle line:90% So I think US is here. 00:30:23.840 --> 00:30:27.510 align:middle line:84% It was very low, and I don't know-- 00:30:27.510 --> 00:30:30.780 align:middle line:84% maybe they couldn't get into China to interview people, so no 00:30:30.780 --> 00:30:32.370 align:middle line:90% China here. 00:30:32.370 --> 00:30:33.730 align:middle line:90% So that's the point. 00:30:33.730 --> 00:30:36.690 align:middle line:90% And then, let's see-- 00:30:36.690 --> 00:30:40.050 align:middle line:84% so as you know-- so let me do an interesting thing. 00:30:40.050 --> 00:30:45.000 align:middle line:84% Now for the entire United States adults, 72%. 00:30:45.000 --> 00:30:48.280 align:middle line:84% Let's do a small poll here in this classroom. 00:30:48.280 --> 00:30:51.240 align:middle line:84% This is not a representative sample. 00:30:51.240 --> 00:30:56.820 align:middle line:84% It's too biased a sample towards some very dark place. 00:30:56.820 --> 00:31:00.030 align:middle line:84% Who thinks global warming is happening, please 00:31:00.030 --> 00:31:02.400 align:middle line:90% raise your hands. 00:31:02.400 --> 00:31:04.440 align:middle line:90% You don't think so. 00:31:04.440 --> 00:31:06.960 align:middle line:90% 00:31:06.960 --> 00:31:07.590 align:middle line:90% OK, fine. 00:31:07.590 --> 00:31:10.560 align:middle line:84% We are not counting-- no pressure. 00:31:10.560 --> 00:31:12.880 align:middle line:84% AUDIENCE: [INAUDIBLE] comments from the-- 00:31:12.880 --> 00:31:13.570 align:middle line:90% SIQI ZHENG: Yes. 00:31:13.570 --> 00:31:15.630 align:middle line:90% So I just want to show this. 00:31:15.630 --> 00:31:21.420 align:middle line:84% If we say United States, 72%, this 11.350-- 00:31:21.420 --> 00:31:22.720 align:middle line:90% that's our classroom. 00:31:22.720 --> 00:31:24.990 align:middle line:90% So [INAUDIBLE] here. 00:31:24.990 --> 00:31:28.120 align:middle line:84% So it's much higher than the national average. 00:31:28.120 --> 00:31:29.400 align:middle line:90% So I think that's a situation. 00:31:29.400 --> 00:31:31.270 align:middle line:90% You know it's heterogeneous. 00:31:31.270 --> 00:31:32.460 align:middle line:90% It's not the same. 00:31:32.460 --> 00:31:33.900 align:middle line:90% You have a comment? 00:31:33.900 --> 00:31:36.760 align:middle line:84% AUDIENCE: I just wanted to comment-- 00:31:36.760 --> 00:31:39.310 align:middle line:90% I just see that Israel is last. 00:31:39.310 --> 00:31:42.310 align:middle line:84% I think the reason for that might be that-- 00:31:42.310 --> 00:31:45.100 align:middle line:84% I can attest for Israel that it's not 00:31:45.100 --> 00:31:46.780 align:middle line:84% that people don't believe in it, it's 00:31:46.780 --> 00:31:50.900 align:middle line:84% just that it doesn't take the place in the media. 00:31:50.900 --> 00:31:51.490 align:middle line:90% I mean, it's-- 00:31:51.490 --> 00:31:52.907 align:middle line:84% SIQI ZHENG: Too many other things. 00:31:52.907 --> 00:31:54.260 align:middle line:90% Yes, that's true. 00:31:54.260 --> 00:31:55.888 align:middle line:84% That's exactly the same as in China. 00:31:55.888 --> 00:31:57.430 align:middle line:84% I think many from China that you know 00:31:57.430 --> 00:32:00.380 align:middle line:84% that there are so many other bigger challenges over there. 00:32:00.380 --> 00:32:04.010 align:middle line:84% The climate is so remote to common people's daily life. 00:32:04.010 --> 00:32:07.780 align:middle line:84% So that's a very good explanation. 00:32:07.780 --> 00:32:10.420 align:middle line:84% And when you are so happy with all other things, 00:32:10.420 --> 00:32:13.300 align:middle line:84% you start to worry about climate, 00:32:13.300 --> 00:32:15.050 align:middle line:84% worry about your next generations. 00:32:15.050 --> 00:32:16.550 align:middle line:84% AUDIENCE: I think it's a good point. 00:32:16.550 --> 00:32:18.552 align:middle line:84% And also, I'd be interested in knowing 00:32:18.552 --> 00:32:20.260 align:middle line:84% the correlation of that list of countries 00:32:20.260 --> 00:32:22.040 align:middle line:90% and beliefs versus impacts. 00:32:22.040 --> 00:32:24.460 align:middle line:84% Like, if you're living somewhere like right on the coast, 00:32:24.460 --> 00:32:27.010 align:middle line:84% if you live in Venice and you're actually feeling it, 00:32:27.010 --> 00:32:30.040 align:middle line:84% versus a landlocked country that maybe you're 00:32:30.040 --> 00:32:31.633 align:middle line:84% not experiencing the impacts as much. 00:32:31.633 --> 00:32:33.550 align:middle line:84% I know coming from California and experiencing 00:32:33.550 --> 00:32:36.113 align:middle line:84% those wildfires, it makes you believe-- 00:32:36.113 --> 00:32:37.780 align:middle line:84% SIQI ZHENG: That your house got burned-- 00:32:37.780 --> 00:32:38.470 align:middle line:90% yes. 00:32:38.470 --> 00:32:39.880 align:middle line:90% AUDIENCE: [INAUDIBLE] happened. 00:32:39.880 --> 00:32:42.562 align:middle line:84% So I'd be interested to see the correlation between that. 00:32:42.562 --> 00:32:43.270 align:middle line:90% SIQI ZHENG: Good. 00:32:43.270 --> 00:32:45.730 align:middle line:84% Yes, that's very good-- another research topic. 00:32:45.730 --> 00:32:48.370 align:middle line:90% And then, see, that's exactly-- 00:32:48.370 --> 00:32:51.430 align:middle line:84% this graph addresses your point just now. 00:32:51.430 --> 00:32:56.050 align:middle line:84% So climate change is a priority, but now it's 00:32:56.050 --> 00:32:58.570 align:middle line:84% more and more partisan than ever. 00:32:58.570 --> 00:33:00.190 align:middle line:90% So they have this table. 00:33:00.190 --> 00:33:02.630 align:middle line:90% They have these trends. 00:33:02.630 --> 00:33:05.170 align:middle line:90% So these tools still-- 00:33:05.170 --> 00:33:07.570 align:middle line:84% I think they're still United States. 00:33:07.570 --> 00:33:10.630 align:middle line:90% And see, four priorities. 00:33:10.630 --> 00:33:14.200 align:middle line:84% Economic growth-- of course, always the first priority. 00:33:14.200 --> 00:33:17.980 align:middle line:90% Jobs-- people want employment. 00:33:17.980 --> 00:33:20.770 align:middle line:84% Environmental protection-- I mean, 00:33:20.770 --> 00:33:22.840 align:middle line:84% like all the environmental things-- 00:33:22.840 --> 00:33:24.200 align:middle line:90% and the global warming. 00:33:24.200 --> 00:33:25.900 align:middle line:84% So this is the percentage of Americans 00:33:25.900 --> 00:33:29.710 align:middle line:84% who say the issue should be a top priority for the president 00:33:29.710 --> 00:33:31.500 align:middle line:90% and the Congress. 00:33:31.500 --> 00:33:34.410 align:middle line:84% So then, clearly, you can see a very interesting thing 00:33:34.410 --> 00:33:38.680 align:middle line:84% is they have cycles, not like they're flat all the time. 00:33:38.680 --> 00:33:43.920 align:middle line:84% Always when the economy is good, then 00:33:43.920 --> 00:33:48.750 align:middle line:84% people are happy with jobs and the wage and all other things, 00:33:48.750 --> 00:33:54.000 align:middle line:84% they have more brain space to worry about climate. 00:33:54.000 --> 00:33:57.660 align:middle line:84% When the economy is so bad, in a recession, 00:33:57.660 --> 00:34:00.420 align:middle line:84% we don't know whether tomorrow we will have lunch or not, 00:34:00.420 --> 00:34:02.570 align:middle line:84% and we are not going to worry about climate. 00:34:02.570 --> 00:34:06.030 align:middle line:84% We need to first secure a job and all the other things. 00:34:06.030 --> 00:34:10.770 align:middle line:84% So you will say this is a financial crisis-- 00:34:10.770 --> 00:34:12.060 align:middle line:90% 2009. 00:34:12.060 --> 00:34:15.302 align:middle line:84% After 2009, the priority-- people 00:34:15.302 --> 00:34:17.260 align:middle line:84% always thought its economic growth and the jobs 00:34:17.260 --> 00:34:18.389 align:middle line:90% and all other things. 00:34:18.389 --> 00:34:20.030 align:middle line:90% It's not about global warming. 00:34:20.030 --> 00:34:26.460 align:middle line:84% And when the economy is becoming better, then they go up. 00:34:26.460 --> 00:34:28.860 align:middle line:84% And then there's Democrats and the Republicans-- 00:34:28.860 --> 00:34:30.940 align:middle line:90% you'll see the gap is huge gap. 00:34:30.940 --> 00:34:32.520 align:middle line:90% Partisan gap-- it's so big. 00:34:32.520 --> 00:34:38.540 align:middle line:84% So that's basically-- and then here is after COVID-19. 00:34:38.540 --> 00:34:39.630 align:middle line:90% This is the recent, 2022. 00:34:39.630 --> 00:34:41.790 align:middle line:90% I got this yesterday. 00:34:41.790 --> 00:34:45.380 align:middle line:84% So now it's a strengthening economy, number one, 00:34:45.380 --> 00:34:51.380 align:middle line:84% because now we are also like very important moment 00:34:51.380 --> 00:34:52.469 align:middle line:90% for the economic growth. 00:34:52.469 --> 00:34:57.050 align:middle line:84% The inflation is so high, so reducing health care costs 00:34:57.050 --> 00:34:58.640 align:middle line:90% or something like that. 00:34:58.640 --> 00:35:00.400 align:middle line:84% And let me see-- there is global warming. 00:35:00.400 --> 00:35:03.390 align:middle line:90% 00:35:03.390 --> 00:35:07.440 align:middle line:84% Yeah, here, it's very small, but global warming is here. 00:35:07.440 --> 00:35:10.560 align:middle line:84% So you can see it's not always a priority, 00:35:10.560 --> 00:35:12.280 align:middle line:90% and it changes over time. 00:35:12.280 --> 00:35:15.160 align:middle line:90% So that's dynamic. 00:35:15.160 --> 00:35:18.260 align:middle line:90% What's the implication of that? 00:35:18.260 --> 00:35:19.650 align:middle line:90% What's the implication? 00:35:19.650 --> 00:35:23.120 align:middle line:84% I'm going to use this paper to illustrate, of course, 00:35:23.120 --> 00:35:26.170 align:middle line:84% we care about public opinions, especially for those 00:35:26.170 --> 00:35:27.920 align:middle line:84% who want to go to the public sector later, 00:35:27.920 --> 00:35:30.170 align:middle line:90% become a policymaker-- sure. 00:35:30.170 --> 00:35:31.800 align:middle line:90% Look at this and study it. 00:35:31.800 --> 00:35:34.340 align:middle line:84% But later, if you go to private sector, 00:35:34.340 --> 00:35:37.980 align:middle line:84% you are just one developer or one small investor, 00:35:37.980 --> 00:35:41.180 align:middle line:84% then you need to understand the big picture to seek 00:35:41.180 --> 00:35:44.200 align:middle line:90% good opportunities in this. 00:35:44.200 --> 00:35:48.600 align:middle line:84% So here, of course, you know climate change is always 00:35:48.600 --> 00:35:49.560 align:middle line:90% on people's mind-- 00:35:49.560 --> 00:35:51.450 align:middle line:84% sometimes very important, sometimes less 00:35:51.450 --> 00:35:53.850 align:middle line:84% important-- in some areas, more important. 00:35:53.850 --> 00:35:56.580 align:middle line:84% Then how that will affect real estate prices? 00:35:56.580 --> 00:35:57.580 align:middle line:90% That's real. 00:35:57.580 --> 00:35:59.580 align:middle line:84% You need to understand, if you go into a market, 00:35:59.580 --> 00:36:05.260 align:middle line:84% you go to the map of the climate belief, check in your submarket 00:36:05.260 --> 00:36:07.260 align:middle line:90% what the belief index is. 00:36:07.260 --> 00:36:11.290 align:middle line:84% And you think about this will affect your business. 00:36:11.290 --> 00:36:13.770 align:middle line:84% That's so real-- will affect your business, 00:36:13.770 --> 00:36:15.550 align:middle line:84% will affect real surprise, later on 00:36:15.550 --> 00:36:18.780 align:middle line:84% will show will affect the housing 00:36:18.780 --> 00:36:22.830 align:middle line:84% developers' profit if you are going to build something 00:36:22.830 --> 00:36:23.940 align:middle line:90% on the coast. 00:36:23.940 --> 00:36:27.720 align:middle line:84% So this paper is the following-- this is a very good journal. 00:36:27.720 --> 00:36:31.800 align:middle line:84% And, of course, they say flood risk is so big, 00:36:31.800 --> 00:36:34.230 align:middle line:84% and especially for Florida and Hawaii, 00:36:34.230 --> 00:36:37.180 align:middle line:90% like, later will be inundated-- 00:36:37.180 --> 00:36:39.070 align:middle line:90% 10% will be inundated. 00:36:39.070 --> 00:36:42.880 align:middle line:84% And then they started, OK, this is scientific fact. 00:36:42.880 --> 00:36:44.670 align:middle line:90% This is scientific report. 00:36:44.670 --> 00:36:47.110 align:middle line:84% Then this paper went into the market, 00:36:47.110 --> 00:36:53.920 align:middle line:84% really got a lot of real transactions from 1997 to 2017 00:36:53.920 --> 00:36:56.320 align:middle line:84% and also detailed housing listings, 00:36:56.320 --> 00:36:58.000 align:middle line:84% whether you are in flood zone or not, 00:36:58.000 --> 00:37:01.090 align:middle line:84% if you are in flood zone, higher risk. 00:37:01.090 --> 00:37:03.780 align:middle line:90% And then that's in Florida. 00:37:03.780 --> 00:37:07.860 align:middle line:84% So underwater means it's a dummy variable that the home is 00:37:07.860 --> 00:37:09.120 align:middle line:90% in the flood zone. 00:37:09.120 --> 00:37:14.100 align:middle line:84% Then they got into this Yale Climate Opinion Map 2016 data. 00:37:14.100 --> 00:37:16.140 align:middle line:90% They ran a hedonic model. 00:37:16.140 --> 00:37:20.730 align:middle line:84% We are all hedonic experts now after my apples-to-apples 00:37:20.730 --> 00:37:23.620 align:middle line:90% comparison lecture last time. 00:37:23.620 --> 00:37:26.230 align:middle line:84% Now we are all apple-to-apple experts. 00:37:26.230 --> 00:37:27.540 align:middle line:90% So we run hedonic-- 00:37:27.540 --> 00:37:29.130 align:middle line:90% we run hedonic. 00:37:29.130 --> 00:37:31.980 align:middle line:84% We regress log P-- that's a housing 00:37:31.980 --> 00:37:36.850 align:middle line:84% price-- on so many other things and all these characteristics. 00:37:36.850 --> 00:37:38.400 align:middle line:90% Then we have underwater. 00:37:38.400 --> 00:37:41.970 align:middle line:90% Underwater means in flood zone. 00:37:41.970 --> 00:37:45.570 align:middle line:84% Then this paper found out the very interesting result. 00:37:45.570 --> 00:37:47.260 align:middle line:90% Always beta is negative. 00:37:47.260 --> 00:37:49.860 align:middle line:84% If you are flood zone, you have a discount. 00:37:49.860 --> 00:37:50.830 align:middle line:90% That's always. 00:37:50.830 --> 00:37:55.140 align:middle line:84% However, the size of the beta varies a lot. 00:37:55.140 --> 00:37:58.680 align:middle line:84% The size of the beta, the discount of the beta, 00:37:58.680 --> 00:38:03.780 align:middle line:84% [? where ?] is allowed for those homes, for those believers, 00:38:03.780 --> 00:38:06.280 align:middle line:84% they call the believers the median. 00:38:06.280 --> 00:38:10.470 align:middle line:84% So they have the Yale Map of the belief that the median value 00:38:10.470 --> 00:38:11.520 align:middle line:90% of the belief-- 00:38:11.520 --> 00:38:15.900 align:middle line:84% about belief, they call this community as believers. 00:38:15.900 --> 00:38:20.520 align:middle line:84% Under the mean value, median value, they call it deniers. 00:38:20.520 --> 00:38:25.500 align:middle line:84% So compare this to the believers community 00:38:25.500 --> 00:38:29.710 align:middle line:84% associated with 7% larger decrease in housing price. 00:38:29.710 --> 00:38:31.450 align:middle line:90% So that's the finding. 00:38:31.450 --> 00:38:36.390 align:middle line:84% This is very intuitive, which means for those for those two-- 00:38:36.390 --> 00:38:39.370 align:middle line:90% remember apple-to-apple-- 00:38:39.370 --> 00:38:43.120 align:middle line:90% 00:38:43.120 --> 00:38:46.150 align:middle line:90% if there are two communities-- 00:38:46.150 --> 00:38:48.940 align:middle line:90% this is the ocean. 00:38:48.940 --> 00:38:52.360 align:middle line:84% These two communities, both on the coast, 00:38:52.360 --> 00:38:57.130 align:middle line:90% the same climate objective risk. 00:38:57.130 --> 00:38:58.780 align:middle line:90% Flooding risk is the same. 00:38:58.780 --> 00:39:00.830 align:middle line:90% These are all believers. 00:39:00.830 --> 00:39:02.890 align:middle line:90% They say this is real. 00:39:02.890 --> 00:39:05.860 align:middle line:90% These are deniers. 00:39:05.860 --> 00:39:09.070 align:middle line:84% The same physical risk-- however, 00:39:09.070 --> 00:39:13.820 align:middle line:84% it says this one's housing price decreases a lot. 00:39:13.820 --> 00:39:16.060 align:middle line:90% This one decreases a little. 00:39:16.060 --> 00:39:18.380 align:middle line:90% So that's what they found. 00:39:18.380 --> 00:39:20.920 align:middle line:90% Then, if you are businessman-- 00:39:20.920 --> 00:39:24.680 align:middle line:90% If you are businessman-- 00:39:24.680 --> 00:39:26.450 align:middle line:90% you are local businessman. 00:39:26.450 --> 00:39:29.350 align:middle line:84% You are going to build something in seaport, for example, 00:39:29.350 --> 00:39:32.320 align:middle line:90% or some Florida coastal area. 00:39:32.320 --> 00:39:35.717 align:middle line:84% And then you only have a few choices. 00:39:35.717 --> 00:39:36.550 align:middle line:90% You have constraint. 00:39:36.550 --> 00:39:38.140 align:middle line:90% You have these few choices-- 00:39:38.140 --> 00:39:39.740 align:middle line:90% very similar physical risk. 00:39:39.740 --> 00:39:42.010 align:middle line:84% And which community are you going 00:39:42.010 --> 00:39:44.205 align:middle line:90% to go to to build your houses? 00:39:44.205 --> 00:39:47.635 align:middle line:90% 00:39:47.635 --> 00:39:49.010 align:middle line:84% AUDIENCE: People who don't agree. 00:39:49.010 --> 00:39:51.740 align:middle line:84% SIQI ZHENG: You go to the deniers. 00:39:51.740 --> 00:39:56.270 align:middle line:84% And do you think for yourself as a businessman, 00:39:56.270 --> 00:39:58.580 align:middle line:90% yes, that's a rational choice. 00:39:58.580 --> 00:39:59.700 align:middle line:90% You get profit. 00:39:59.700 --> 00:40:04.070 align:middle line:84% But if we aggregate 1,000 of businessmen like you, 00:40:04.070 --> 00:40:06.880 align:middle line:84% all go there to build-- is there any problem for the society? 00:40:06.880 --> 00:40:10.272 align:middle line:90% 00:40:10.272 --> 00:40:13.465 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:40:13.465 --> 00:40:14.340 align:middle line:90% SIQI ZHENG: Not sure. 00:40:14.340 --> 00:40:17.440 align:middle line:90% 00:40:17.440 --> 00:40:21.130 align:middle line:84% So if you are an individual business, of course, 00:40:21.130 --> 00:40:25.120 align:middle line:84% go here to build because you'll have higher price, 00:40:25.120 --> 00:40:29.090 align:middle line:84% you'll make profit, and then you have a lot of buyers. 00:40:29.090 --> 00:40:34.120 align:middle line:84% But if you have so many builders all go here and build, 00:40:34.120 --> 00:40:39.970 align:middle line:84% then we'll be more and more housing in riskier areas. 00:40:39.970 --> 00:40:42.940 align:middle line:84% Once there is a hurricane damage, 00:40:42.940 --> 00:40:45.340 align:middle line:84% the damage will be borne by whom? 00:40:45.340 --> 00:40:49.450 align:middle line:84% The damage will be-- because you build, you sell, and you leave. 00:40:49.450 --> 00:40:52.600 align:middle line:84% You go to another place to build. 00:40:52.600 --> 00:40:55.870 align:middle line:84% Who will bear the damage if a lot of buildings 00:40:55.870 --> 00:40:57.720 align:middle line:84% just build here because there are deniers? 00:40:57.720 --> 00:41:00.340 align:middle line:90% 00:41:00.340 --> 00:41:04.620 align:middle line:90% Any sons who will bear the cost. 00:41:04.620 --> 00:41:05.997 align:middle line:90% Who will cry? 00:41:05.997 --> 00:41:07.830 align:middle line:84% AUDIENCE: Residents-- people who live there. 00:41:07.830 --> 00:41:09.288 align:middle line:84% SIQI ZHENG: The residents will cry. 00:41:09.288 --> 00:41:12.118 align:middle line:90% And who else will cry? 00:41:12.118 --> 00:41:13.540 align:middle line:90% AUDIENCE: Insurance companies. 00:41:13.540 --> 00:41:14.320 align:middle line:90% SIQI ZHENG: Yes. 00:41:14.320 --> 00:41:17.370 align:middle line:90% And who else? 00:41:17.370 --> 00:41:19.810 align:middle line:90% AUDIENCE: The deniers. 00:41:19.810 --> 00:41:21.310 align:middle line:84% SIQI ZHENG: OK, they now understand. 00:41:21.310 --> 00:41:22.240 align:middle line:90% They learned. 00:41:22.240 --> 00:41:26.380 align:middle line:84% And also banks-- the banks because they buy these houses 00:41:26.380 --> 00:41:30.670 align:middle line:84% using mortgage and then all the people then get inundated, 00:41:30.670 --> 00:41:34.090 align:middle line:84% then they default. The households will default. 00:41:34.090 --> 00:41:35.455 align:middle line:90% Take over my house-- 00:41:35.455 --> 00:41:37.150 align:middle line:90% I default. Remember? 00:41:37.150 --> 00:41:39.970 align:middle line:90% Then banks will cry. 00:41:39.970 --> 00:41:43.370 align:middle line:84% And these are low-income household-- they couldn't leave. 00:41:43.370 --> 00:41:46.870 align:middle line:84% They have to leave there, and they will cry. 00:41:46.870 --> 00:41:48.980 align:middle line:90% So that's the social problem. 00:41:48.980 --> 00:41:51.080 align:middle line:84% So I just want to keep this open. 00:41:51.080 --> 00:41:53.060 align:middle line:84% And later, you can discuss it a lot. 00:41:53.060 --> 00:41:57.070 align:middle line:84% So basically, you can see all these interesting topics. 00:41:57.070 --> 00:42:00.760 align:middle line:84% And then, of course, now we understand-- 00:42:00.760 --> 00:42:02.830 align:middle line:84% not so many people they know Yale-- 00:42:02.830 --> 00:42:05.320 align:middle line:84% that Yale University map-- before, I just mentioned, 00:42:05.320 --> 00:42:07.615 align:middle line:90% who of you know-- 00:42:07.615 --> 00:42:11.620 align:middle line:84% except our teaching people, we all know. 00:42:11.620 --> 00:42:17.730 align:middle line:84% Who knew this climate belief map existed? 00:42:17.730 --> 00:42:18.820 align:middle line:90% You've heard of that? 00:42:18.820 --> 00:42:20.320 align:middle line:84% AUDIENCE: I kind of knew that that's 00:42:20.320 --> 00:42:23.640 align:middle line:84% where [INAUDIBLE] would believe it based on politics, 00:42:23.640 --> 00:42:26.798 align:middle line:84% like what [INAUDIBLE] said, Democrats more likely to believe 00:42:26.798 --> 00:42:30.190 align:middle line:90% that climate change exists. 00:42:30.190 --> 00:42:32.020 align:middle line:90% SIQI ZHENG: Yes. 00:42:32.020 --> 00:42:38.860 align:middle line:84% So that means, at least, we cannot force people to believe 00:42:38.860 --> 00:42:41.570 align:middle line:84% something or not believe something. 00:42:41.570 --> 00:42:45.830 align:middle line:84% They have their freedom of belief in climate change or not. 00:42:45.830 --> 00:42:47.060 align:middle line:90% That's their freedom. 00:42:47.060 --> 00:42:51.160 align:middle line:84% But I think, as a private sector or the government, 00:42:51.160 --> 00:42:53.200 align:middle line:90% they should provide information. 00:42:53.200 --> 00:42:56.500 align:middle line:84% They should provide information for people 00:42:56.500 --> 00:42:57.740 align:middle line:90% to judge by themselves. 00:42:57.740 --> 00:42:59.630 align:middle line:90% Instead, they don't understand. 00:42:59.630 --> 00:43:02.650 align:middle line:84% They even don't know their houses are in flood zone. 00:43:02.650 --> 00:43:04.460 align:middle line:90% That's not good. 00:43:04.460 --> 00:43:08.420 align:middle line:90% So this disclosure is important. 00:43:08.420 --> 00:43:11.290 align:middle line:84% This is very similar, but it's the opposite example 00:43:11.290 --> 00:43:13.750 align:middle line:84% of the green building certificate, remember? 00:43:13.750 --> 00:43:15.190 align:middle line:84% When I talk about green buildings, 00:43:15.190 --> 00:43:16.600 align:middle line:90% I talk about market failure. 00:43:16.600 --> 00:43:19.360 align:middle line:84% One market failure is information asymmetry. 00:43:19.360 --> 00:43:23.000 align:middle line:84% So people don't know whether they see a green building 00:43:23.000 --> 00:43:27.140 align:middle line:84% or not like that, and then LEED, Energy Star, 00:43:27.140 --> 00:43:30.080 align:middle line:84% and all those certificate-- of course, 00:43:30.080 --> 00:43:33.020 align:middle line:84% they have their own problems and limitations, 00:43:33.020 --> 00:43:35.810 align:middle line:84% but at least they made a big contribution 00:43:35.810 --> 00:43:38.570 align:middle line:84% to solve the information asymmetry problem. 00:43:38.570 --> 00:43:40.200 align:middle line:90% That is green building. 00:43:40.200 --> 00:43:41.630 align:middle line:90% This is a negative thing. 00:43:41.630 --> 00:43:44.990 align:middle line:84% Climate risk-- also need the information. 00:43:44.990 --> 00:43:46.820 align:middle line:90% Provide information. 00:43:46.820 --> 00:43:51.020 align:middle line:84% So that's why now they have this Home Seller Disclosure 00:43:51.020 --> 00:43:52.970 align:middle line:90% Requirement-- 00:43:52.970 --> 00:43:55.910 align:middle line:84% mandate the sellers, no matter the individual sellers 00:43:55.910 --> 00:43:59.660 align:middle line:84% or developers, provide buyers with detailed information 00:43:59.660 --> 00:44:02.400 align:middle line:90% of the known material defects. 00:44:02.400 --> 00:44:04.220 align:middle line:84% It's not just about climate, but climate 00:44:04.220 --> 00:44:07.800 align:middle line:84% is a new item they require to put into that disclosure. 00:44:07.800 --> 00:44:11.360 align:middle line:84% At first, it's disclose all other things-- 00:44:11.360 --> 00:44:16.560 align:middle line:84% sewage and this and basement and building materials like that. 00:44:16.560 --> 00:44:20.760 align:middle line:84% Now they say, now it's time to add climate risk 00:44:20.760 --> 00:44:22.800 align:middle line:90% into that disclosure file. 00:44:22.800 --> 00:44:25.440 align:middle line:84% When you buy a house, you need to-- when 00:44:25.440 --> 00:44:27.570 align:middle line:84% the buyer buys the house, you need to tell them. 00:44:27.570 --> 00:44:33.210 align:middle line:84% And then this mandate was implemented step by step, 00:44:33.210 --> 00:44:34.170 align:middle line:90% not immediately. 00:44:34.170 --> 00:44:36.960 align:middle line:84% It's very hard to get things immediately. 00:44:36.960 --> 00:44:40.440 align:middle line:84% Then some states started doing this, 00:44:40.440 --> 00:44:42.400 align:middle line:90% and some states lag behind. 00:44:42.400 --> 00:44:44.040 align:middle line:84% You can say this also correlation 00:44:44.040 --> 00:44:47.940 align:middle line:84% with that partisan or belief map. 00:44:47.940 --> 00:44:50.580 align:middle line:90% 00:44:50.580 --> 00:44:53.430 align:middle line:90% Here quickly, and this slowly-- 00:44:53.430 --> 00:44:55.740 align:middle line:90% but they are doing so. 00:44:55.740 --> 00:45:00.796 align:middle line:84% And my lab's postdoc, [? Xinghong-- ?] some of you 00:45:00.796 --> 00:45:03.060 align:middle line:84% know [? Xinghong-- ?] [? Xinghong ?] is supposed to be 00:45:03.060 --> 00:45:04.680 align:middle line:90% my lab. 00:45:04.680 --> 00:45:10.800 align:middle line:84% He has a very good paper to study this disclosure's impact 00:45:10.800 --> 00:45:17.010 align:middle line:90% on housing market outcomes. 00:45:17.010 --> 00:45:23.290 align:middle line:84% And he said, OK, the home seller disclosure requirement, 00:45:23.290 --> 00:45:26.260 align:middle line:90% this better information helps. 00:45:26.260 --> 00:45:30.480 align:middle line:84% Then, he showed that in those states and counties 00:45:30.480 --> 00:45:33.690 align:middle line:84% with a disclosure implemented, then 00:45:33.690 --> 00:45:38.070 align:middle line:84% you see this place is losing people and higher housing 00:45:38.070 --> 00:45:38.910 align:middle line:90% vacancy. 00:45:38.910 --> 00:45:43.920 align:middle line:84% So when people know this place is risky, they start to leave, 00:45:43.920 --> 00:45:46.530 align:middle line:84% and then the market starts to decline. 00:45:46.530 --> 00:45:48.610 align:middle line:84% So that's the information channel. 00:45:48.610 --> 00:45:56.520 align:middle line:84% You just give information and let people judge by themselves. 00:45:56.520 --> 00:46:00.030 align:middle line:84% OK, that's [? Xinghong's ?] research. 00:46:00.030 --> 00:46:02.710 align:middle line:90% And my lab has another research. 00:46:02.710 --> 00:46:03.690 align:middle line:90% We have a website-- 00:46:03.690 --> 00:46:04.840 align:middle line:90% search Google. 00:46:04.840 --> 00:46:06.210 align:middle line:90% We have a climate-- 00:46:06.210 --> 00:46:08.500 align:middle line:84% we have a global sentiment project. 00:46:08.500 --> 00:46:10.320 align:middle line:84% So this project is the following-- 00:46:10.320 --> 00:46:12.030 align:middle line:90% also linked to climate change. 00:46:12.030 --> 00:46:14.980 align:middle line:84% So we use natural language processing. 00:46:14.980 --> 00:46:17.160 align:middle line:84% That's a kind of machine learning-based way 00:46:17.160 --> 00:46:19.920 align:middle line:84% to extract information from social media. 00:46:19.920 --> 00:46:23.410 align:middle line:84% We started from Weibo, China's Twitter. 00:46:23.410 --> 00:46:26.520 align:middle line:84% Then we expand it to all Twitter, 00:46:26.520 --> 00:46:28.650 align:middle line:84% So all over the world-- now we have 100-- more 00:46:28.650 --> 00:46:30.330 align:middle line:90% than 100 countries. 00:46:30.330 --> 00:46:32.250 align:middle line:84% And this is machine learning, so we can 00:46:32.250 --> 00:46:34.000 align:middle line:90% process all kinds of languages. 00:46:34.000 --> 00:46:36.580 align:middle line:90% We don't rely on dictionary. 00:46:36.580 --> 00:46:42.750 align:middle line:84% So then we use this NLP approach to extract information 00:46:42.750 --> 00:46:44.830 align:middle line:90% about people's sentiment. 00:46:44.830 --> 00:46:48.332 align:middle line:84% Twitter, Tweets-- whether you are happy or not. 00:46:48.332 --> 00:46:50.040 align:middle line:84% We have a sentiment score for each tweet, 00:46:50.040 --> 00:46:51.810 align:middle line:90% and then we aggregate. 00:46:51.810 --> 00:46:58.960 align:middle line:84% And then we link the sentiment to these extreme temperatures 00:46:58.960 --> 00:47:03.640 align:middle line:84% to show that when it's very hot or it's very cold, 00:47:03.640 --> 00:47:05.300 align:middle line:90% people are less happy. 00:47:05.300 --> 00:47:06.710 align:middle line:90% It seems very intuitive. 00:47:06.710 --> 00:47:09.247 align:middle line:90% But it's really heterogeneous. 00:47:09.247 --> 00:47:11.080 align:middle line:84% You can see different countries because they 00:47:11.080 --> 00:47:13.890 align:middle line:84% have different adaptation strategies. 00:47:13.890 --> 00:47:16.490 align:middle line:84% It's not all the countries are rich, as in the United States. 00:47:16.490 --> 00:47:18.730 align:middle line:84% You have air conditioning, you have heating. 00:47:18.730 --> 00:47:21.310 align:middle line:84% No, some countries-- very hard to 00:47:21.310 --> 00:47:24.100 align:middle line:90% adapt to extreme temperatures. 00:47:24.100 --> 00:47:26.840 align:middle line:84% And then we have different countries. 00:47:26.840 --> 00:47:31.060 align:middle line:84% Then we quantify-- we call it emotional toll, the climate 00:47:31.060 --> 00:47:34.030 align:middle line:84% change's emotional toll on people. 00:47:34.030 --> 00:47:37.070 align:middle line:84% And basically, that's why you see the inverse U. 00:47:37.070 --> 00:47:39.320 align:middle line:90% So this is our paper. 00:47:39.320 --> 00:47:41.710 align:middle line:90% We will soon publish it. 00:47:41.710 --> 00:47:43.360 align:middle line:84% And then, of course, you can link that 00:47:43.360 --> 00:47:45.080 align:middle line:90% to all kinds of other outcomes. 00:47:45.080 --> 00:47:47.110 align:middle line:90% Actually, we already link to-- 00:47:47.110 --> 00:47:48.880 align:middle line:84% Juan and me and the [INAUDIBLE] paper 00:47:48.880 --> 00:47:51.640 align:middle line:84% will link to COVID the same sentiment 00:47:51.640 --> 00:47:54.970 align:middle line:84% and also air pollution and other negative shocks. 00:47:54.970 --> 00:47:56.820 align:middle line:90% So this is our website. 00:47:56.820 --> 00:48:01.340 align:middle line:84% Go there and take a look of our projects and the team 00:48:01.340 --> 00:48:02.980 align:middle line:90% if you want to use this index. 00:48:02.980 --> 00:48:06.350 align:middle line:90% 00:48:06.350 --> 00:48:09.740 align:middle line:84% The third project for my lab-- related. 00:48:09.740 --> 00:48:13.550 align:middle line:84% And it's also the information belief matters. 00:48:13.550 --> 00:48:17.270 align:middle line:84% So this is an ongoing project by my PhD student, [INAUDIBLE]. 00:48:17.270 --> 00:48:20.990 align:middle line:84% And so, basically, this says-- this 00:48:20.990 --> 00:48:24.110 align:middle line:84% is very interesting-- we further explore 00:48:24.110 --> 00:48:26.360 align:middle line:84% Twitter, not just are they happy or not, 00:48:26.360 --> 00:48:28.490 align:middle line:90% but also their friendship. 00:48:28.490 --> 00:48:31.242 align:middle line:84% You know, on Twitter, you have followers and followees-- 00:48:31.242 --> 00:48:33.450 align:middle line:84% followers, and you follow others-- others follow you. 00:48:33.450 --> 00:48:35.520 align:middle line:90% We know that information. 00:48:35.520 --> 00:48:40.010 align:middle line:84% So we create the friendship network, 00:48:40.010 --> 00:48:41.600 align:middle line:90% which means your friends-- 00:48:41.600 --> 00:48:45.260 align:middle line:84% not just here-- of course, here, all of us, we are friends. 00:48:45.260 --> 00:48:46.260 align:middle line:90% We are good friends. 00:48:46.260 --> 00:48:50.030 align:middle line:84% But you also have friends in Harvard and another state 00:48:50.030 --> 00:48:51.600 align:middle line:90% and another country like that. 00:48:51.600 --> 00:48:53.840 align:middle line:90% You have your social network. 00:48:53.840 --> 00:48:56.770 align:middle line:90% Then we say this-- 00:48:56.770 --> 00:49:02.210 align:middle line:84% like Hurricane Ida just hit Louisiana. 00:49:02.210 --> 00:49:05.180 align:middle line:84% In Louisiana, people suddenly got so unhappy 00:49:05.180 --> 00:49:07.580 align:middle line:84% because they lost homes and like this-- 00:49:07.580 --> 00:49:08.760 align:middle line:90% so unhappy. 00:49:08.760 --> 00:49:10.580 align:middle line:84% And then they chatted on Twitter. 00:49:10.580 --> 00:49:12.920 align:middle line:84% And this information not just in Louisiana-- 00:49:12.920 --> 00:49:16.680 align:middle line:84% more space got hit and also spread out. 00:49:16.680 --> 00:49:21.060 align:middle line:84% So Louisiana first, and then Mississippi and Rhode island 00:49:21.060 --> 00:49:23.010 align:middle line:90% and California like that. 00:49:23.010 --> 00:49:27.480 align:middle line:84% So this is the response, which means 00:49:27.480 --> 00:49:31.140 align:middle line:84% the hurricane hit one place, and then, because of social network, 00:49:31.140 --> 00:49:32.790 align:middle line:90% that's a social learning. 00:49:32.790 --> 00:49:36.030 align:middle line:84% And then the friend here, here, here, here, they 00:49:36.030 --> 00:49:38.320 align:middle line:90% learned, oh, this is dangerous. 00:49:38.320 --> 00:49:41.970 align:middle line:84% And at first, they didn't know what hurricane mean because they 00:49:41.970 --> 00:49:43.200 align:middle line:90% never experienced it. 00:49:43.200 --> 00:49:45.450 align:middle line:84% Then, suddenly, the pictures on Twitter 00:49:45.450 --> 00:49:50.360 align:middle line:84% shows my friend in Louisiana got hit and so bad, so miserable. 00:49:50.360 --> 00:49:51.850 align:middle line:90% Then they understand the risk. 00:49:51.850 --> 00:49:55.020 align:middle line:84% Then they adjust their belief and all the things. 00:49:55.020 --> 00:49:58.500 align:middle line:84% So that's basically our matrix of the information-- hurricane 00:49:58.500 --> 00:50:01.320 align:middle line:84% information networks, which say we 00:50:01.320 --> 00:50:04.050 align:middle line:84% have more friend with the other states 00:50:04.050 --> 00:50:06.180 align:middle line:90% and how this information flow. 00:50:06.180 --> 00:50:08.110 align:middle line:90% So that's another ongoing thing. 00:50:08.110 --> 00:50:13.000 align:middle line:84% So these are all about beliefs, so it's very interesting. 00:50:13.000 --> 00:50:14.430 align:middle line:90% It's a very soft thing. 00:50:14.430 --> 00:50:16.260 align:middle line:90% It's not very concrete. 00:50:16.260 --> 00:50:20.680 align:middle line:84% But we can measure and we understand that really matters. 00:50:20.680 --> 00:50:24.150 align:middle line:84% And the soft thing will lead to tangible things 00:50:24.150 --> 00:50:28.240 align:middle line:90% of the new development. 00:50:28.240 --> 00:50:30.760 align:middle line:90% Now we are going to go move on. 00:50:30.760 --> 00:50:32.120 align:middle line:90% Belief is important. 00:50:32.120 --> 00:50:35.980 align:middle line:84% Now let's see how to quantify how 00:50:35.980 --> 00:50:39.310 align:middle line:84% to quantify the impact of the climate risk 00:50:39.310 --> 00:50:41.950 align:middle line:90% on housing outcomes. 00:50:41.950 --> 00:50:46.300 align:middle line:84% So this paper is one of the first papers I think 00:50:46.300 --> 00:50:48.680 align:middle line:90% that's on your reading list. 00:50:48.680 --> 00:50:50.120 align:middle line:90% And this is a very good paper. 00:50:50.120 --> 00:50:51.953 align:middle line:84% It's a very good journal, a very good paper, 00:50:51.953 --> 00:50:55.180 align:middle line:84% one of the first several papers, and very easy to understand 00:50:55.180 --> 00:50:55.900 align:middle line:90% paper-- 00:50:55.900 --> 00:50:58.940 align:middle line:84% Hurricane Sandy and New York City housing market. 00:50:58.940 --> 00:51:02.530 align:middle line:84% So I will use this, but I will expand. 00:51:02.530 --> 00:51:06.160 align:middle line:84% I will tell a big story in the methodology in this paper. 00:51:06.160 --> 00:51:09.520 align:middle line:84% And after you understand this paper, you will become expert. 00:51:09.520 --> 00:51:12.550 align:middle line:84% And then you can easily do yourself-- 00:51:12.550 --> 00:51:16.810 align:middle line:84% I know you are not looking at a seaport in your studio project. 00:51:16.810 --> 00:51:22.930 align:middle line:84% And that one is very high risk of the sea level rise. 00:51:22.930 --> 00:51:27.010 align:middle line:84% So you can use our data and this methodology 00:51:27.010 --> 00:51:31.010 align:middle line:84% to quantify the climate risk on your seaport project. 00:51:31.010 --> 00:51:33.350 align:middle line:84% I think that's something later we 00:51:33.350 --> 00:51:37.240 align:middle line:84% will need to do as our P set number 2 and some analysis 00:51:37.240 --> 00:51:40.480 align:middle line:90% in Carlos' class. 00:51:40.480 --> 00:51:45.150 align:middle line:84% Now I'm going to enter the boring part, abstract part. 00:51:45.150 --> 00:51:48.700 align:middle line:84% So hedonic model on climate risks-- 00:51:48.700 --> 00:51:52.100 align:middle line:84% the thing is clearly, for example, Hurricane Sandy-- 00:51:52.100 --> 00:51:55.010 align:middle line:90% 00:51:55.010 --> 00:51:57.230 align:middle line:84% so, for example, there's a hurricane. 00:51:57.230 --> 00:52:01.250 align:middle line:84% The hurricane hit an area, a city. 00:52:01.250 --> 00:52:04.080 align:middle line:84% Then, of course, there's a flood zone. 00:52:04.080 --> 00:52:06.050 align:middle line:90% And outside the flood zone-- 00:52:06.050 --> 00:52:08.930 align:middle line:84% for now, we assume the flood zone 00:52:08.930 --> 00:52:10.540 align:middle line:90% is very accurately measured. 00:52:10.540 --> 00:52:13.400 align:middle line:84% Actually, it is not, but for now, we assume. 00:52:13.400 --> 00:52:16.010 align:middle line:84% This is a flood zone-- just get flooded. 00:52:16.010 --> 00:52:18.200 align:middle line:84% And the other zones, they are not 00:52:18.200 --> 00:52:20.600 align:middle line:84% flooded because the flood zone may 00:52:20.600 --> 00:52:24.190 align:middle line:84% be very low, elevation like that, and very close 00:52:24.190 --> 00:52:26.240 align:middle line:90% to the coastline. 00:52:26.240 --> 00:52:31.740 align:middle line:84% Then we say, OK, we want to quantify this flood risk 00:52:31.740 --> 00:52:34.620 align:middle line:90% impact on housing prices. 00:52:34.620 --> 00:52:36.920 align:middle line:84% Now we have a very simple way to do so. 00:52:36.920 --> 00:52:37.920 align:middle line:90% We compare. 00:52:37.920 --> 00:52:41.460 align:middle line:84% We compare this "treated"-- we call the "treated" 00:52:41.460 --> 00:52:44.820 align:middle line:90% this one got flooded with this-- 00:52:44.820 --> 00:52:46.020 align:middle line:90% without flooding. 00:52:46.020 --> 00:52:50.220 align:middle line:84% Then we calculate the gap of the housing price between these two 00:52:50.220 --> 00:52:53.370 align:middle line:84% areas, and then we draw a conclusion. 00:52:53.370 --> 00:52:56.702 align:middle line:90% Is there any problem on this? 00:52:56.702 --> 00:52:57.810 align:middle line:90% [INAUDIBLE] here? 00:52:57.810 --> 00:53:00.450 align:middle line:84% AUDIENCE: [INAUDIBLE] it's not [INAUDIBLE]. 00:53:00.450 --> 00:53:01.970 align:middle line:90% SIQI ZHENG: Good. 00:53:01.970 --> 00:53:05.100 align:middle line:84% Now we understand this is not apple-to-apple. 00:53:05.100 --> 00:53:09.720 align:middle line:84% So today, I don't have real apples and oranges. 00:53:09.720 --> 00:53:11.910 align:middle line:84% I didn't have time to go to Trader Joe's yesterday, 00:53:11.910 --> 00:53:13.680 align:middle line:90% but next time I will bring more. 00:53:13.680 --> 00:53:16.980 align:middle line:90% But now let's see-- 00:53:16.980 --> 00:53:18.450 align:middle line:90% just now. 00:53:18.450 --> 00:53:21.720 align:middle line:84% I already said we cannot just compare all these [INAUDIBLE] 00:53:21.720 --> 00:53:23.870 align:middle line:90% and calculate the price gap. 00:53:23.870 --> 00:53:25.080 align:middle line:90% That's too naive. 00:53:25.080 --> 00:53:28.010 align:middle line:84% Then it's all apples-to-oranges comparison. 00:53:28.010 --> 00:53:30.410 align:middle line:84% But could you elaborate a little bit more? 00:53:30.410 --> 00:53:33.450 align:middle line:84% What kind of things potentially are different between these two 00:53:33.450 --> 00:53:35.000 align:middle line:90% zones? 00:53:35.000 --> 00:53:37.130 align:middle line:84% We are talking about apple is not orange. 00:53:37.130 --> 00:53:38.530 align:middle line:90% That doesn't equal to an orange. 00:53:38.530 --> 00:53:41.520 align:middle line:90% What are potential differences? 00:53:41.520 --> 00:53:47.040 align:middle line:84% AUDIENCE: [INAUDIBLE] properties on [INAUDIBLE] might be, like, 00:53:47.040 --> 00:53:48.960 align:middle line:90% [INAUDIBLE] 00:53:48.960 --> 00:53:51.570 align:middle line:84% SIQI ZHENG: Why the control zone will be older? 00:53:51.570 --> 00:53:53.134 align:middle line:90% What's the rationale? 00:53:53.134 --> 00:53:57.874 align:middle line:84% AUDIENCE: Maybe it's [INAUDIBLE] that 00:53:57.874 --> 00:53:59.780 align:middle line:84% have been developed for a long time 00:53:59.780 --> 00:54:04.570 align:middle line:84% and the treated zone is the [INAUDIBLE] 00:54:04.570 --> 00:54:07.060 align:middle line:84% SIQI ZHENG: You are thinking about the Seaport, 00:54:07.060 --> 00:54:10.090 align:middle line:84% the control is like Boston downtown. 00:54:10.090 --> 00:54:13.420 align:middle line:84% And then they expand and then become the coastal area. 00:54:13.420 --> 00:54:14.510 align:middle line:90% OK, that's possible. 00:54:14.510 --> 00:54:15.085 align:middle line:90% Any others? 00:54:15.085 --> 00:54:19.200 align:middle line:90% 00:54:19.200 --> 00:54:21.580 align:middle line:84% What do you think is a systematic difference? 00:54:21.580 --> 00:54:26.370 align:middle line:84% We must look at the systematic, not random difference. 00:54:26.370 --> 00:54:28.680 align:middle line:90% AUDIENCE: The one-- excuse me-- 00:54:28.680 --> 00:54:31.140 align:middle line:84% some of the ones that could be in the flood zone 00:54:31.140 --> 00:54:33.690 align:middle line:84% or affected by hurricane could have a premium on them 00:54:33.690 --> 00:54:36.810 align:middle line:84% because of preferred views or better location 00:54:36.810 --> 00:54:41.460 align:middle line:84% versus some control zone may be more inland commuter locations 00:54:41.460 --> 00:54:44.252 align:middle line:90% versus premier locations. 00:54:44.252 --> 00:54:45.710 align:middle line:84% SIQI ZHENG: You are talking about-- 00:54:45.710 --> 00:54:48.470 align:middle line:84% actually, this treated ones may be very good 00:54:48.470 --> 00:54:50.660 align:middle line:90% will because ocean-- 00:54:50.660 --> 00:54:52.490 align:middle line:90% everyone likes ocean view. 00:54:52.490 --> 00:54:56.750 align:middle line:84% So if you like ocean view, you live in on beach, 00:54:56.750 --> 00:55:00.990 align:middle line:84% you're on the beach, you need to bear the risk of being flooded. 00:55:00.990 --> 00:55:02.660 align:middle line:90% So that's one difference. 00:55:02.660 --> 00:55:03.673 align:middle line:90% Any other things? 00:55:03.673 --> 00:55:05.840 align:middle line:84% AUDIENCE: There could also be the flip side of that, 00:55:05.840 --> 00:55:08.730 align:middle line:84% like maybe inside the flood zone, 00:55:08.730 --> 00:55:11.150 align:middle line:84% it was a poorer neighborhood, and maybe there 00:55:11.150 --> 00:55:14.290 align:middle line:84% wasn't as much investment and red lining and different things 00:55:14.290 --> 00:55:16.420 align:middle line:84% like that, so it wasn't as protected 00:55:16.420 --> 00:55:18.140 align:middle line:90% as outside of the flood zone. 00:55:18.140 --> 00:55:19.340 align:middle line:90% SIQI ZHENG: Good, yes. 00:55:19.340 --> 00:55:23.320 align:middle line:84% Maybe because the inside the flood zone already declining-- 00:55:23.320 --> 00:55:25.270 align:middle line:84% declining neighborhood and people 00:55:25.270 --> 00:55:27.490 align:middle line:84% understood that, then people left because we just 00:55:27.490 --> 00:55:28.720 align:middle line:90% showed paper, people left-- 00:55:28.720 --> 00:55:32.920 align:middle line:84% [INAUDIBLE] paper-- people left, and not enough property tax 00:55:32.920 --> 00:55:35.420 align:middle line:84% revenue, then very badly maintained-- 00:55:35.420 --> 00:55:37.570 align:middle line:90% so they're already bad. 00:55:37.570 --> 00:55:39.230 align:middle line:84% So these are all the different things. 00:55:39.230 --> 00:55:45.070 align:middle line:84% And another thing is flood insurance because it's required, 00:55:45.070 --> 00:55:47.800 align:middle line:84% although it's not very well-enforced. 00:55:47.800 --> 00:55:50.560 align:middle line:84% But it's required if you borrow mortgage from the bank, 00:55:50.560 --> 00:55:54.460 align:middle line:84% if you are in a flood zone, you need to buy flood insurance. 00:55:54.460 --> 00:55:56.380 align:middle line:90% So they are insured. 00:55:56.380 --> 00:55:58.640 align:middle line:90% And these are not insured. 00:55:58.640 --> 00:56:02.140 align:middle line:84% So that will be like, oh, I'm insured, so not-- no problem. 00:56:02.140 --> 00:56:04.130 align:middle line:90% Even if I get hit-- insured. 00:56:04.130 --> 00:56:06.410 align:middle line:90% So that is a difference. 00:56:06.410 --> 00:56:10.220 align:middle line:90% 00:56:10.220 --> 00:56:11.870 align:middle line:90% Think about this. 00:56:11.870 --> 00:56:16.340 align:middle line:84% Last time, I used this to help you better understand-- in 00:56:16.340 --> 00:56:19.830 align:middle line:84% a very numerical way understand this. 00:56:19.830 --> 00:56:21.410 align:middle line:90% Let's look at this. 00:56:21.410 --> 00:56:26.000 align:middle line:84% Let's think about the problem of that apple-to-orange comparison. 00:56:26.000 --> 00:56:29.430 align:middle line:84% So, for example, there is a house, $1 million. 00:56:29.430 --> 00:56:31.430 align:middle line:90% Then we decompose that hedonic-- 00:56:31.430 --> 00:56:32.900 align:middle line:90% decompose. 00:56:32.900 --> 00:56:35.990 align:middle line:84% Three rooms, two busways, two subway stops, 00:56:35.990 --> 00:56:38.690 align:middle line:84% and a nice park and some trash can-- 00:56:38.690 --> 00:56:40.070 align:middle line:90% the trash can here. 00:56:40.070 --> 00:56:43.310 align:middle line:84% And now we have that inside flood zone-- 00:56:43.310 --> 00:56:46.700 align:middle line:84% now we know inside flood zone bad. 00:56:46.700 --> 00:56:50.150 align:middle line:84% Then, after a lot of data, crunching 00:56:50.150 --> 00:56:54.320 align:middle line:84% the data and the regression things, negative 0.1 million 00:56:54.320 --> 00:56:55.190 align:middle line:90% discount. 00:56:55.190 --> 00:56:58.970 align:middle line:90% So that's the hedonic model. 00:56:58.970 --> 00:57:03.410 align:middle line:84% Thus, we cannot compare no matter apple/orange you put 00:57:03.410 --> 00:57:04.880 align:middle line:90% together to compare. 00:57:04.880 --> 00:57:07.130 align:middle line:84% Then this basically what we [INAUDIBLE] I 00:57:07.130 --> 00:57:08.960 align:middle line:84% think that our [INAUDIBLE] number two. 00:57:08.960 --> 00:57:10.660 align:middle line:84% Keep in mind this kind of thing-- 00:57:10.660 --> 00:57:13.320 align:middle line:90% we are interested in alpha 3. 00:57:13.320 --> 00:57:14.390 align:middle line:90% This is a trash can. 00:57:14.390 --> 00:57:17.450 align:middle line:84% That's why we have an R square, which is not 100%. 00:57:17.450 --> 00:57:19.890 align:middle line:90% It is 3%. 00:57:19.890 --> 00:57:21.750 align:middle line:90% So that's the idea. 00:57:21.750 --> 00:57:23.990 align:middle line:84% I think you have the intuition of this. 00:57:23.990 --> 00:57:26.730 align:middle line:90% 00:57:26.730 --> 00:57:28.710 align:middle line:90% Now, clearly, we understand. 00:57:28.710 --> 00:57:30.610 align:middle line:90% The trash can is not that clean. 00:57:30.610 --> 00:57:32.060 align:middle line:90% It's not random. 00:57:32.060 --> 00:57:34.500 align:middle line:84% There are some things in the trash can 00:57:34.500 --> 00:57:40.860 align:middle line:84% which may be correlated with this flood zone location. 00:57:40.860 --> 00:57:42.900 align:middle line:90% So now I have one example-- 00:57:42.900 --> 00:57:46.320 align:middle line:84% to make this simple one example, which is ocean view. 00:57:46.320 --> 00:57:50.130 align:middle line:84% As we said, the ocean seaport is very easy 00:57:50.130 --> 00:57:52.410 align:middle line:90% that you are in a flood zone. 00:57:52.410 --> 00:57:55.045 align:middle line:84% Now the trash can is not that clean. 00:57:55.045 --> 00:57:57.660 align:middle line:90% 00:57:57.660 --> 00:58:00.360 align:middle line:84% So if we can observe with a measure-- 00:58:00.360 --> 00:58:03.030 align:middle line:84% for example, we are so sophisticated we can measure 00:58:03.030 --> 00:58:04.565 align:middle line:90% exactly the ocean view's value. 00:58:04.565 --> 00:58:07.510 align:middle line:90% 00:58:07.510 --> 00:58:09.170 align:middle line:90% That's a hypothetical scenario. 00:58:09.170 --> 00:58:11.940 align:middle line:90% But anyway, if we can observe-- 00:58:11.940 --> 00:58:14.820 align:middle line:90% suppose it's $0.05 million. 00:58:14.820 --> 00:58:16.860 align:middle line:84% And then this will be the remaining trash 00:58:16.860 --> 00:58:21.810 align:middle line:84% can or something we really want to measure, like that. 00:58:21.810 --> 00:58:26.290 align:middle line:90% Then we are very confident. 00:58:26.290 --> 00:58:30.600 align:middle line:84% But if we cannot observe the ocean view 00:58:30.600 --> 00:58:36.840 align:middle line:84% and if, in the extreme case, that the ocean view means flood 00:58:36.840 --> 00:58:38.940 align:middle line:84% zone-- these two are the same thing-- 00:58:38.940 --> 00:58:41.490 align:middle line:90% then you have a trouble. 00:58:41.490 --> 00:58:43.275 align:middle line:90% So, basically, you will see-- 00:58:43.275 --> 00:58:46.780 align:middle line:90% 00:58:46.780 --> 00:58:51.630 align:middle line:84% I think this is the-- let me say ocean view. 00:58:51.630 --> 00:58:53.430 align:middle line:84% So basically, you will say here-- 00:58:53.430 --> 00:58:58.250 align:middle line:84% so ocean view will go into here because ocean view and the flood 00:58:58.250 --> 00:58:59.610 align:middle line:90% zone is the same thing. 00:58:59.610 --> 00:59:02.340 align:middle line:84% So you naively run the regression. 00:59:02.340 --> 00:59:04.320 align:middle line:84% You don't know these two are different. 00:59:04.320 --> 00:59:05.720 align:middle line:90% You think that's the same. 00:59:05.720 --> 00:59:07.610 align:middle line:90% Then you get this. 00:59:07.610 --> 00:59:10.500 align:middle line:84% Then you underestimate the discount. 00:59:10.500 --> 00:59:13.700 align:middle line:84% You say, OK, in the flood zone, now it's not a big deal. 00:59:13.700 --> 00:59:18.470 align:middle line:84% In the flood zone, you only have a discount of $0.05 million, 00:59:18.470 --> 00:59:21.710 align:middle line:84% because in the flood zone, you also 00:59:21.710 --> 00:59:23.430 align:middle line:84% have a benefit of the ocean view. 00:59:23.430 --> 00:59:29.540 align:middle line:84% Actually, if you can really split the ocean view out, 00:59:29.540 --> 00:59:31.880 align:middle line:84% actually, and the flood zone is so bad 00:59:31.880 --> 00:59:36.110 align:middle line:84% that's negative $0.1 million, then this one 00:59:36.110 --> 00:59:38.820 align:middle line:84% offset this bad thing, this negative amenity. 00:59:38.820 --> 00:59:40.220 align:middle line:90% This is a positive. 00:59:40.220 --> 00:59:42.930 align:middle line:84% If you can really measure, you get this right. 00:59:42.930 --> 00:59:45.270 align:middle line:90% If you cannot, you get it wrong. 00:59:45.270 --> 00:59:47.100 align:middle line:90% You underestimate. 00:59:47.100 --> 00:59:52.420 align:middle line:84% So that's the big challenge of this kind of research. 00:59:52.420 --> 00:59:55.410 align:middle line:84% We call it identification because it's identification 00:59:55.410 --> 00:59:58.470 align:middle line:84% challenge is this all kinds of things in the trash 00:59:58.470 --> 01:00:02.730 align:middle line:84% can get correlated with the thing you are mostly interested 01:00:02.730 --> 01:00:04.350 align:middle line:90% in. 01:00:04.350 --> 01:00:08.580 align:middle line:84% Don't worry-- we have a way to handle this 01:00:08.580 --> 01:00:10.770 align:middle line:90% to make this to apple-to-apple. 01:00:10.770 --> 01:00:15.570 align:middle line:84% So this is called difference-in-difference, DID. 01:00:15.570 --> 01:00:17.790 align:middle line:90% So maybe you heard of this term. 01:00:17.790 --> 01:00:20.490 align:middle line:84% Anyway, I think [INAUDIBLE] last time 01:00:20.490 --> 01:00:23.820 align:middle line:84% you mentioned you know DID, right? 01:00:23.820 --> 01:00:25.770 align:middle line:84% Can you use a very simple language 01:00:25.770 --> 01:00:28.700 align:middle line:84% to explain to us what is the idea about? 01:00:28.700 --> 01:00:34.745 align:middle line:90% 01:00:34.745 --> 01:00:36.370 align:middle line:84% AUDIENCE: I've got to think about that. 01:00:36.370 --> 01:00:37.630 align:middle line:90% SIQI ZHENG: OK, you forgot. 01:00:37.630 --> 01:00:38.690 align:middle line:90% OK, no problem. 01:00:38.690 --> 01:00:39.960 align:middle line:90% Take your time. 01:00:39.960 --> 01:00:41.650 align:middle line:90% AUDIENCE: It's more like. 01:00:41.650 --> 01:00:44.860 align:middle line:84% Instead of you directly try to find a relation 01:00:44.860 --> 01:00:47.360 align:middle line:84% shape between two [INAUDIBLE] stuff, 01:00:47.360 --> 01:00:49.100 align:middle line:84% you try to find like an intermediate 01:00:49.100 --> 01:00:51.690 align:middle line:90% as a vehicle in the middle. 01:00:51.690 --> 01:00:53.130 align:middle line:90% And you-- 01:00:53.130 --> 01:00:55.700 align:middle line:84% SIQI ZHENG: Oh, that's an instrumental variable. 01:00:55.700 --> 01:00:57.652 align:middle line:90% AUDIENCE: Oh, OK 01:00:57.652 --> 01:00:59.360 align:middle line:84% SIQI ZHENG: So anyway, now we are talking 01:00:59.360 --> 01:01:01.200 align:middle line:90% about different methodologies. 01:01:01.200 --> 01:01:04.760 align:middle line:84% Remember, last time, in my first apple-to-apple comparison, 01:01:04.760 --> 01:01:08.060 align:middle line:84% I taught a methodology called what? 01:01:08.060 --> 01:01:12.857 align:middle line:84% I hope you can remember the terminology. 01:01:12.857 --> 01:01:14.690 align:middle line:84% AUDIENCE: Sorry, I actually don't, but I can 01:01:14.690 --> 01:01:16.905 align:middle line:90% list [INAUDIBLE] all of them. 01:01:16.905 --> 01:01:17.780 align:middle line:90% SIQI ZHENG: Matching. 01:01:17.780 --> 01:01:18.572 align:middle line:90% AUDIENCE: Matching. 01:01:18.572 --> 01:01:19.410 align:middle line:90% SIQI ZHENG: Yes. 01:01:19.410 --> 01:01:21.310 align:middle line:90% The simple way is matching. 01:01:21.310 --> 01:01:25.380 align:middle line:84% Remember, last time, we talked about matching. 01:01:25.380 --> 01:01:30.750 align:middle line:84% So all kinds of ways for us to have apple-to-apple. 01:01:30.750 --> 01:01:33.180 align:middle line:84% Last time, we talked green building-- that's matching. 01:01:33.180 --> 01:01:40.450 align:middle line:84% Green-- nongreen-- very close by. 01:01:40.450 --> 01:01:42.990 align:middle line:84% Then we match green with nongreen. 01:01:42.990 --> 01:01:46.490 align:middle line:84% The same location, similar things, all similar-- 01:01:46.490 --> 01:01:48.200 align:middle line:90% we calculate the difference. 01:01:48.200 --> 01:01:49.505 align:middle line:90% This is called matching. 01:01:49.505 --> 01:01:52.380 align:middle line:90% 01:01:52.380 --> 01:01:54.540 align:middle line:84% Now this is the first methodology 01:01:54.540 --> 01:01:56.100 align:middle line:90% I hope you can grasp. 01:01:56.100 --> 01:01:58.560 align:middle line:84% Now today, the second one is DID. 01:01:58.560 --> 01:02:01.720 align:middle line:84% DID I'm going to use a very intuitive way. 01:02:01.720 --> 01:02:05.070 align:middle line:84% I'm going to show so that you can understand. 01:02:05.070 --> 01:02:08.440 align:middle line:84% Now just now, we already said they are different animals-- 01:02:08.440 --> 01:02:11.230 align:middle line:84% one is apple, the other is orange. 01:02:11.230 --> 01:02:12.790 align:middle line:90% We cannot direct compare. 01:02:12.790 --> 01:02:15.080 align:middle line:90% That will lead to bias. 01:02:15.080 --> 01:02:17.860 align:middle line:90% We show bias. 01:02:17.860 --> 01:02:19.820 align:middle line:90% Now we have a way-- 01:02:19.820 --> 01:02:23.510 align:middle line:84% fortunately, this is a over time thing. 01:02:23.510 --> 01:02:25.980 align:middle line:84% It's not like we only observe one time point. 01:02:25.980 --> 01:02:28.100 align:middle line:90% We observe over time. 01:02:28.100 --> 01:02:31.460 align:middle line:84% So we observe-- like Sandy, this hurricane-- 01:02:31.460 --> 01:02:35.210 align:middle line:90% before Sandy and after Sandy. 01:02:35.210 --> 01:02:36.980 align:middle line:90% So we take advantage of that. 01:02:36.980 --> 01:02:41.180 align:middle line:84% If you can only observe one time point, that's all, 01:02:41.180 --> 01:02:42.210 align:middle line:90% then you cannot do. 01:02:42.210 --> 01:02:45.940 align:middle line:84% If you observe all the time, you have this way to do. 01:02:45.940 --> 01:02:48.700 align:middle line:90% 01:02:48.700 --> 01:02:51.610 align:middle line:90% Before Sandy is a smaller apple. 01:02:51.610 --> 01:02:54.070 align:middle line:84% I use this [INAUDIBLE] so always, 01:02:54.070 --> 01:02:55.420 align:middle line:90% the housing price go up-- 01:02:55.420 --> 01:02:56.810 align:middle line:90% in normal time, go up. 01:02:56.810 --> 01:03:00.010 align:middle line:84% So before Sandy, the apple is smaller. 01:03:00.010 --> 01:03:02.420 align:middle line:90% Then it grow larger naturally-- 01:03:02.420 --> 01:03:04.630 align:middle line:90% naturally grow larger. 01:03:04.630 --> 01:03:07.790 align:middle line:90% And can we just compare-- 01:03:07.790 --> 01:03:12.700 align:middle line:84% no question-- can we just compare this small apple 01:03:12.700 --> 01:03:15.832 align:middle line:84% and this big apple-- say, OK, I understand I cannot compare 01:03:15.832 --> 01:03:16.540 align:middle line:90% apple and orange. 01:03:16.540 --> 01:03:18.340 align:middle line:90% But now I'm going to change. 01:03:18.340 --> 01:03:21.460 align:middle line:84% I compare before Sandy, what's the price? 01:03:21.460 --> 01:03:23.080 align:middle line:90% After sandy, what's the price? 01:03:23.080 --> 01:03:24.400 align:middle line:90% Subtract. 01:03:24.400 --> 01:03:26.620 align:middle line:84% Then this is the impact of Sandy. 01:03:26.620 --> 01:03:28.560 align:middle line:84% Any problem of this or it's good? 01:03:28.560 --> 01:03:32.980 align:middle line:90% 01:03:32.980 --> 01:03:33.640 align:middle line:90% Go ahead. 01:03:33.640 --> 01:03:34.390 align:middle line:84% AUDIENCE: I mean, something might 01:03:34.390 --> 01:03:36.932 align:middle line:84% have-- something else might have changed in the neighborhood. 01:03:36.932 --> 01:03:38.050 align:middle line:90% SIQI ZHENG: For example? 01:03:38.050 --> 01:03:39.340 align:middle line:90% AUDIENCE: I don't know-- 01:03:39.340 --> 01:03:45.830 align:middle line:84% like new retail in the area [INAUDIBLE] home prices up. 01:03:45.830 --> 01:03:47.572 align:middle line:90% SIQI ZHENG: OK, good. 01:03:47.572 --> 01:03:50.080 align:middle line:84% AUDIENCE: [INAUDIBLE] market appreciation. 01:03:50.080 --> 01:03:53.700 align:middle line:84% I mean, that also doesn't factor in here. 01:03:53.700 --> 01:03:56.700 align:middle line:84% SIQI ZHENG: Like subway stops-- like something like that. 01:03:56.700 --> 01:03:58.080 align:middle line:90% Now it's naive. 01:03:58.080 --> 01:03:59.880 align:middle line:90% This direction-- naive. 01:03:59.880 --> 01:04:02.220 align:middle line:90% This direction-- naive. 01:04:02.220 --> 01:04:07.020 align:middle line:84% Now we have seen this control before-- 01:04:07.020 --> 01:04:12.250 align:middle line:84% small orange to big orange also grow over time. 01:04:12.250 --> 01:04:14.280 align:middle line:84% Now this is called difference-in-difference 01:04:14.280 --> 01:04:14.885 align:middle line:90% in the way. 01:04:14.885 --> 01:04:18.020 align:middle line:90% 01:04:18.020 --> 01:04:22.630 align:middle line:84% First, the steps-- first, measure the baseline difference. 01:04:22.630 --> 01:04:24.930 align:middle line:84% Baseline means before the treatment. 01:04:24.930 --> 01:04:26.570 align:middle line:90% That's the baseline. 01:04:26.570 --> 01:04:29.300 align:middle line:84% Measure the baseline difference between pre-apple 01:04:29.300 --> 01:04:30.240 align:middle line:90% and pre-orange. 01:04:30.240 --> 01:04:31.660 align:middle line:90% So we call it pretreatment. 01:04:31.660 --> 01:04:33.180 align:middle line:84% Actually, it's called pre-treatment. 01:04:33.180 --> 01:04:37.310 align:middle line:84% But I use this language called pre-apple and pre-orange. 01:04:37.310 --> 01:04:38.400 align:middle line:90% So there's small ones. 01:04:38.400 --> 01:04:40.218 align:middle line:84% So let's measure this baseline difference. 01:04:40.218 --> 01:04:41.010 align:middle line:90% They are different. 01:04:41.010 --> 01:04:42.940 align:middle line:90% So we get this. 01:04:42.940 --> 01:04:45.540 align:middle line:90% 01:04:45.540 --> 01:04:49.500 align:middle line:90% Second, this is [INAUDIBLE]. 01:04:49.500 --> 01:04:51.960 align:middle line:84% This is-- they are different animals. 01:04:51.960 --> 01:04:54.840 align:middle line:84% Orange and apple-- totally different-- 01:04:54.840 --> 01:04:56.530 align:middle line:84% nutrition and all things different. 01:04:56.530 --> 01:05:02.910 align:middle line:84% Then subtract the baseline difference from the underline 01:05:02.910 --> 01:05:06.510 align:middle line:90% after treatment-- 01:05:06.510 --> 01:05:09.100 align:middle line:84% difference between this post-apple and post-orange 01:05:09.100 --> 01:05:10.580 align:middle line:84% to get the real treatment effect. 01:05:10.580 --> 01:05:11.590 align:middle line:90% So that's the point. 01:05:11.590 --> 01:05:15.880 align:middle line:84% First, measure before anything has happened. 01:05:15.880 --> 01:05:18.790 align:middle line:84% What's the fundamental difference of the baseline? 01:05:18.790 --> 01:05:21.850 align:middle line:90% The two small orange and apple. 01:05:21.850 --> 01:05:24.810 align:middle line:90% Then we know they are different. 01:05:24.810 --> 01:05:25.780 align:middle line:90% We control for that. 01:05:25.780 --> 01:05:28.810 align:middle line:84% Now, after treatment, they become more different, 01:05:28.810 --> 01:05:30.380 align:middle line:90% and then we subtract. 01:05:30.380 --> 01:05:33.910 align:middle line:90% So that's the hurricane effect. 01:05:33.910 --> 01:05:37.180 align:middle line:90% And another thing is to this. 01:05:37.180 --> 01:05:38.630 align:middle line:90% This is another way. 01:05:38.630 --> 01:05:40.310 align:middle line:90% It's the same. 01:05:40.310 --> 01:05:43.250 align:middle line:84% First, measure the baseline time trend 01:05:43.250 --> 01:05:44.720 align:middle line:90% from pre-orange to post-orange. 01:05:44.720 --> 01:05:48.510 align:middle line:84% This thing hurricane never happened in this neighborhood. 01:05:48.510 --> 01:05:51.080 align:middle line:84% No hurricane thing never happened. 01:05:51.080 --> 01:05:54.150 align:middle line:90% General trend-- general trend. 01:05:54.150 --> 01:05:57.170 align:middle line:90% Go from small to big-- 01:05:57.170 --> 01:05:59.150 align:middle line:90% measure this. 01:05:59.150 --> 01:06:03.640 align:middle line:84% Then subtract this natural growth 01:06:03.640 --> 01:06:06.970 align:middle line:90% from the treatment growth. 01:06:06.970 --> 01:06:10.870 align:middle line:84% Then you will find that this and this are the same, 01:06:10.870 --> 01:06:12.220 align:middle line:90% You just rearrange. 01:06:12.220 --> 01:06:13.670 align:middle line:90% You just rearrange. 01:06:13.670 --> 01:06:17.340 align:middle line:90% 01:06:17.340 --> 01:06:19.260 align:middle line:90% So each way is fine. 01:06:19.260 --> 01:06:20.200 align:middle line:90% Either way is fine. 01:06:20.200 --> 01:06:23.550 align:middle line:84% But you understand that you have four cells, 01:06:23.550 --> 01:06:28.020 align:middle line:84% and then you first get some difference, then subtract. 01:06:28.020 --> 01:06:30.790 align:middle line:84% That's why it's called difference-in-differences. 01:06:30.790 --> 01:06:34.300 align:middle line:84% This is the two differences-- difference. 01:06:34.300 --> 01:06:36.510 align:middle line:90% So that's the DID. 01:06:36.510 --> 01:06:37.920 align:middle line:90% It's a very intuitive. 01:06:37.920 --> 01:06:40.110 align:middle line:90% Any questions here? 01:06:40.110 --> 01:06:41.500 align:middle line:90% I need to pause a little bit. 01:06:41.500 --> 01:06:44.520 align:middle line:90% 01:06:44.520 --> 01:06:47.650 align:middle line:84% Are you fully comprehend?-- it's not that easy. 01:06:47.650 --> 01:06:50.425 align:middle line:90% 01:06:50.425 --> 01:06:52.050 align:middle line:84% AUDIENCE: On the left side, how similar 01:06:52.050 --> 01:06:53.630 align:middle line:90% does the left side need to be? 01:06:53.630 --> 01:06:58.020 align:middle line:84% I mean, there's already a [INAUDIBLE]. 01:06:58.020 --> 01:06:59.770 align:middle line:84% SIQI ZHENG: Yes, that's a very good point. 01:06:59.770 --> 01:07:03.592 align:middle line:84% So basically, the thing is this fundamental-- 01:07:03.592 --> 01:07:05.550 align:middle line:84% I think you are asking a fundamental assumption 01:07:05.550 --> 01:07:07.110 align:middle line:90% of this methodology. 01:07:07.110 --> 01:07:10.150 align:middle line:84% To make sure this methodology holds, 01:07:10.150 --> 01:07:13.690 align:middle line:84% we have a fundamental assumption is 01:07:13.690 --> 01:07:17.020 align:middle line:84% except for the treatment we are interested in, which is Sandy, 01:07:17.020 --> 01:07:24.910 align:middle line:84% other natural things grows are similar between this and that. 01:07:24.910 --> 01:07:29.820 align:middle line:84% The only different thing is Hurricane Sandy. 01:07:29.820 --> 01:07:33.760 align:middle line:84% Besides Hurricane Sandy, all the macroeconomy and weather 01:07:33.760 --> 01:07:38.050 align:middle line:84% there is some booming of the city 01:07:38.050 --> 01:07:43.350 align:middle line:84% and other things are the same between this orange and apple. 01:07:43.350 --> 01:07:50.050 align:middle line:84% So, actually, we are using this orange natural growth 01:07:50.050 --> 01:07:52.540 align:middle line:84% to infer-- this orange natural growth 01:07:52.540 --> 01:07:57.640 align:middle line:84% to infer for this treatment group 01:07:57.640 --> 01:08:01.750 align:middle line:84% if hurricane hadn't happened, what would have 01:08:01.750 --> 01:08:04.220 align:middle line:90% been for this treatment group? 01:08:04.220 --> 01:08:06.650 align:middle line:90% That's a counterfactual. 01:08:06.650 --> 01:08:08.190 align:middle line:90% We cannot know that. 01:08:08.190 --> 01:08:10.770 align:middle line:84% Nobody can know because it already happened. 01:08:10.770 --> 01:08:16.720 align:middle line:84% But we believe these treatment and control, 01:08:16.720 --> 01:08:19.010 align:middle line:84% in the absence of that treatment, 01:08:19.010 --> 01:08:20.700 align:middle line:90% will follow the same trend. 01:08:20.700 --> 01:08:26.410 align:middle line:84% So we use this trend to infer the trend 01:08:26.410 --> 01:08:31.710 align:middle line:84% of this in the absence of that treatment. 01:08:31.710 --> 01:08:34.270 align:middle line:84% So that's the counterfactual thing. 01:08:34.270 --> 01:08:35.950 align:middle line:90% This is the counterfactual here. 01:08:35.950 --> 01:08:37.680 align:middle line:90% So that's the assumption. 01:08:37.680 --> 01:08:40.590 align:middle line:84% We must test that assumption is true. 01:08:40.590 --> 01:08:44.340 align:middle line:84% AUDIENCE: Everything else is called constant [INAUDIBLE] then 01:08:44.340 --> 01:08:47.399 align:middle line:84% if the building material is subpar on this side, 01:08:47.399 --> 01:08:49.560 align:middle line:90% and then it corrodes faster-- 01:08:49.560 --> 01:08:51.920 align:middle line:84% so do we still need to control characteristics? 01:08:51.920 --> 01:08:52.890 align:middle line:84% SIQI ZHENG: Yeah, we can control. 01:08:52.890 --> 01:08:54.348 align:middle line:84% We can control as much as possible. 01:08:54.348 --> 01:08:56.548 align:middle line:84% But for the trash can, we assume they are similar. 01:08:56.548 --> 01:08:58.840 align:middle line:84% AUDIENCE: Since they're controlling for characteristics 01:08:58.840 --> 01:09:00.965 align:middle line:84% so much, what's the difference between diff-in-diff 01:09:00.965 --> 01:09:01.680 align:middle line:90% and matching? 01:09:01.680 --> 01:09:03.059 align:middle line:90% How different is it? 01:09:03.059 --> 01:09:05.790 align:middle line:84% SIQI ZHENG: Oh, they are used in different scenarios. 01:09:05.790 --> 01:09:08.390 align:middle line:84% So for this over time matching, you 01:09:08.390 --> 01:09:10.340 align:middle line:84% don't need over time [INAUDIBLE] if you only 01:09:10.340 --> 01:09:13.250 align:middle line:84% observe one cross-sectional thing, you can still do. 01:09:13.250 --> 01:09:17.000 align:middle line:84% This one, maybe you only have two areas, 01:09:17.000 --> 01:09:19.200 align:middle line:84% but you don't have so many together. 01:09:19.200 --> 01:09:20.189 align:middle line:90% Then you use this. 01:09:20.189 --> 01:09:23.729 align:middle line:84% They are not like one is much better than the other. 01:09:23.729 --> 01:09:26.640 align:middle line:84% It's totally data and all the context. 01:09:26.640 --> 01:09:30.120 align:middle line:84% AUDIENCE: I would imagine that immediately after the hurricane, 01:09:30.120 --> 01:09:33.370 align:middle line:84% the growth in the apple would be a lot smaller. 01:09:33.370 --> 01:09:36.870 align:middle line:84% But say if you extended the study over a much longer period 01:09:36.870 --> 01:09:40.636 align:middle line:84% of time, say 6-7 years, maybe that difference would start to-- 01:09:40.636 --> 01:09:41.594 align:middle line:90% SIQI ZHENG: To diverge? 01:09:41.594 --> 01:09:44.910 align:middle line:84% AUDIENCE: Yeah, it would not be so much-- 01:09:44.910 --> 01:09:45.785 align:middle line:90% SIQI ZHENG: Yes, yes. 01:09:45.785 --> 01:09:48.748 align:middle line:84% AUDIENCE: I'm curious to know what the period of the study is? 01:09:48.748 --> 01:09:50.290 align:middle line:84% SIQI ZHENG: Yes, that's a good point. 01:09:50.290 --> 01:09:53.220 align:middle line:90% It's a definite answer. 01:09:53.220 --> 01:09:55.930 align:middle line:84% That depends on the data and the study research question. 01:09:55.930 --> 01:10:00.220 align:middle line:84% But it's not like yesterday and today-- not like that. 01:10:00.220 --> 01:10:03.040 align:middle line:84% It's not so immediate change because all the-- 01:10:03.040 --> 01:10:06.360 align:middle line:84% housing markets gradually change, so years, 01:10:06.360 --> 01:10:10.440 align:middle line:84% like two years before, two years after, something like that. 01:10:10.440 --> 01:10:11.040 align:middle line:90% Yeah? 01:10:11.040 --> 01:10:17.024 align:middle line:84% AUDIENCE: So the [INAUDIBLE] between the DID [INAUDIBLE] 01:10:17.024 --> 01:10:20.317 align:middle line:90% climate risk. 01:10:20.317 --> 01:10:22.400 align:middle line:84% SIQI ZHENG: Yeah, that's the impact of the climate 01:10:22.400 --> 01:10:24.270 align:middle line:90% risk on the outcome. 01:10:24.270 --> 01:10:25.980 align:middle line:84% The outcome may be a housing price. 01:10:25.980 --> 01:10:28.880 align:middle line:84% That's why the treatment is the climate. 01:10:28.880 --> 01:10:30.260 align:middle line:90% OK, good. 01:10:30.260 --> 01:10:33.320 align:middle line:90% OK, now let's position. 01:10:33.320 --> 01:10:34.930 align:middle line:90% Now remember this. 01:10:34.930 --> 01:10:37.820 align:middle line:84% This is a framework that you don't need to memorize. 01:10:37.820 --> 01:10:40.255 align:middle line:90% If you grasp all this-- 01:10:40.255 --> 01:10:42.860 align:middle line:90% 01:10:42.860 --> 01:10:43.890 align:middle line:90% PhD graduate. 01:10:43.890 --> 01:10:46.290 align:middle line:90% So don't worry. 01:10:46.290 --> 01:10:48.410 align:middle line:84% I need to first make sure that you don't worry. 01:10:48.410 --> 01:10:54.770 align:middle line:84% But the idea is, as I remember, as I said last time, we really 01:10:54.770 --> 01:10:58.700 align:middle line:84% hope you can grasp all of [INAUDIBLE] run regression 01:10:58.700 --> 01:11:03.830 align:middle line:84% and you have some more sophisticated specification. 01:11:03.830 --> 01:11:06.560 align:middle line:90% But you really want to do this-- 01:11:06.560 --> 01:11:08.600 align:middle line:90% you say, OK, now I understand. 01:11:08.600 --> 01:11:10.830 align:middle line:84% Causal inference is so important. 01:11:10.830 --> 01:11:14.340 align:middle line:84% We just don't want to draw naive conclusion, just a correlation. 01:11:14.340 --> 01:11:17.360 align:middle line:84% If you have that goal in your mind, 01:11:17.360 --> 01:11:23.240 align:middle line:84% you want to be more advanced, then we already talked matching. 01:11:23.240 --> 01:11:25.580 align:middle line:84% Matching, like geographically matching, like that, 01:11:25.580 --> 01:11:28.640 align:middle line:84% and more sophisticated called propensity 01:11:28.640 --> 01:11:32.570 align:middle line:84% score matching-- no need to really get into this unless you 01:11:32.570 --> 01:11:34.070 align:middle line:90% want to do a PhD. 01:11:34.070 --> 01:11:36.410 align:middle line:84% And now, today, we are talking about DID. 01:11:36.410 --> 01:11:38.480 align:middle line:84% Difference-in-differences is DID. 01:11:38.480 --> 01:11:40.380 align:middle line:90% So that's the thing here. 01:11:40.380 --> 01:11:41.990 align:middle line:90% So now we have at least two. 01:11:41.990 --> 01:11:46.700 align:middle line:84% We grasp two out of four instrumental variables-- 01:11:46.700 --> 01:11:48.290 align:middle line:90% so sophisticated. 01:11:48.290 --> 01:11:50.850 align:middle line:84% RDD-- later, I will touch a little bit on this. 01:11:50.850 --> 01:11:53.850 align:middle line:84% And then you know the golden thing is the experiment. 01:11:53.850 --> 01:11:56.700 align:middle line:84% But not so everyone is lucky to run an experiment. 01:11:56.700 --> 01:11:59.280 align:middle line:84% But if you do an experiment, that's the best. 01:11:59.280 --> 01:12:02.160 align:middle line:84% So now you understand the big picture. 01:12:02.160 --> 01:12:06.030 align:middle line:84% And then, to give you some algebra-- 01:12:06.030 --> 01:12:07.230 align:middle line:90% it's so easy. 01:12:07.230 --> 01:12:12.430 align:middle line:84% So this property inside the flood zone. 01:12:12.430 --> 01:12:15.040 align:middle line:84% This property outside the flood zone-- this. 01:12:15.040 --> 01:12:16.660 align:middle line:90% So that's over time. 01:12:16.660 --> 01:12:20.060 align:middle line:84% Now I [INAUDIBLE] this is economist bad-- 01:12:20.060 --> 01:12:21.760 align:middle line:84% so overall, this trend is going down, 01:12:21.760 --> 01:12:23.180 align:middle line:90% but many times, it's going up. 01:12:23.180 --> 01:12:25.570 align:middle line:84% But anyway, let's say the overall trend-- 01:12:25.570 --> 01:12:26.890 align:middle line:90% going down. 01:12:26.890 --> 01:12:32.290 align:middle line:84% Then we understand all other trends similar. 01:12:32.290 --> 01:12:34.180 align:middle line:90% Suddenly, there is a hurricane. 01:12:34.180 --> 01:12:37.640 align:middle line:84% Then, for the treatment, will get a bigger jump. 01:12:37.640 --> 01:12:43.150 align:middle line:84% So our focus, our goal is to get this treatment effect, which 01:12:43.150 --> 01:12:46.090 align:middle line:90% is this big gap right here. 01:12:46.090 --> 01:12:49.630 align:middle line:84% Then the orange thing is the algebra is like this-- 01:12:49.630 --> 01:12:53.245 align:middle line:84% treated control yt1, t2, c1, c2-- 01:12:53.245 --> 01:12:55.120 align:middle line:84% then you just have the first difference, then 01:12:55.120 --> 01:12:56.037 align:middle line:90% the second difference. 01:12:56.037 --> 01:13:00.980 align:middle line:84% So that's the expression of this. 01:13:00.980 --> 01:13:06.260 align:middle line:84% And then just now we call it parallel trend is this-- 01:13:06.260 --> 01:13:10.790 align:middle line:84% before these two groups should have the similar trend? 01:13:10.790 --> 01:13:12.120 align:middle line:90% That's the assumption. 01:13:12.120 --> 01:13:16.110 align:middle line:84% If this assumption doesn't hold, then game over-- 01:13:16.110 --> 01:13:17.230 align:middle line:90% cannot use. 01:13:17.230 --> 01:13:19.670 align:middle line:84% So to do the DID [INAUDIBLE] the first test, 01:13:19.670 --> 01:13:22.730 align:middle line:84% the parallel trend is before anything has happened, 01:13:22.730 --> 01:13:25.380 align:middle line:84% these two have the similar trend. 01:13:25.380 --> 01:13:29.470 align:middle line:84% This is called parallel trend assumption. 01:13:29.470 --> 01:13:32.710 align:middle line:84% If they do not have this, you cannot use, 01:13:32.710 --> 01:13:35.680 align:middle line:84% but not at the end of word because we have other 01:13:35.680 --> 01:13:36.810 align:middle line:90% approaches. 01:13:36.810 --> 01:13:38.900 align:middle line:90% Oh, DID cannot be used-- 01:13:38.900 --> 01:13:40.720 align:middle line:90% then we go to other. 01:13:40.720 --> 01:13:44.160 align:middle line:84% That's econometricians-- they always daily dealing 01:13:44.160 --> 01:13:44.910 align:middle line:90% with these things. 01:13:44.910 --> 01:13:48.280 align:middle line:90% 01:13:48.280 --> 01:13:50.480 align:middle line:84% Now I talk too much about methodology. 01:13:50.480 --> 01:13:53.790 align:middle line:90% Let's see some application. 01:13:53.790 --> 01:13:57.080 align:middle line:90% So I'm going to have this. 01:13:57.080 --> 01:13:58.700 align:middle line:90% I go back to that paper. 01:13:58.700 --> 01:14:00.260 align:middle line:90% That paper you will read. 01:14:00.260 --> 01:14:04.670 align:middle line:90% So Sandy, 2012, hit New York. 01:14:04.670 --> 01:14:07.460 align:middle line:84% And then they use it-- they say, OK, this 01:14:07.460 --> 01:14:10.730 align:middle line:84% is a perfect setting for DID because not all 01:14:10.730 --> 01:14:12.000 align:middle line:90% the communities got hit. 01:14:12.000 --> 01:14:13.920 align:middle line:90% Some got, some not-- 01:14:13.920 --> 01:14:17.060 align:middle line:90% then treatment and control. 01:14:17.060 --> 01:14:20.330 align:middle line:84% And then the question is Hurricane Sandy on New York City 01:14:20.330 --> 01:14:21.060 align:middle line:90% housing market. 01:14:21.060 --> 01:14:22.990 align:middle line:90% What's the impact of that? 01:14:22.990 --> 01:14:25.330 align:middle line:84% And then they say, OK, they collect all the data. 01:14:25.330 --> 01:14:26.950 align:middle line:84% They knew New York City is not that difficult to collect 01:14:26.950 --> 01:14:28.220 align:middle line:90% the housing transaction. 01:14:28.220 --> 01:14:29.260 align:middle line:90% It's very easy. 01:14:29.260 --> 01:14:31.030 align:middle line:90% And then they geocode. 01:14:31.030 --> 01:14:34.540 align:middle line:84% And then also they got FEMA data on the damage. 01:14:34.540 --> 01:14:37.000 align:middle line:84% The good thing is they really have the damage. 01:14:37.000 --> 01:14:39.190 align:middle line:84% They have the damage data, the building-- they 01:14:39.190 --> 01:14:42.280 align:middle line:90% have several damage levels. 01:14:42.280 --> 01:14:44.110 align:middle line:84% They find properties in flood zone 01:14:44.110 --> 01:14:47.970 align:middle line:84% has 8% persistent price penalty after Sandy. 01:14:47.970 --> 01:14:54.280 align:middle line:84% So remember, 8%, persistent-- not like after Sandy the next 01:14:54.280 --> 01:14:55.520 align:middle line:90% year, 8%-- 01:14:55.520 --> 01:14:57.730 align:middle line:84% the second year, people forgot about that 01:14:57.730 --> 01:15:02.180 align:middle line:84% and then recover, and then later fully recover. 01:15:02.180 --> 01:15:04.870 align:middle line:90% They said no, persistent. 01:15:04.870 --> 01:15:06.080 align:middle line:90% So that's your finding. 01:15:06.080 --> 01:15:07.660 align:middle line:90% But I need to let you know. 01:15:07.660 --> 01:15:11.950 align:middle line:84% Now this area of research-- a lot of papers, and they 01:15:11.950 --> 01:15:13.360 align:middle line:90% have different findings. 01:15:13.360 --> 01:15:15.050 align:middle line:90% They are inconsistent. 01:15:15.050 --> 01:15:18.980 align:middle line:84% Some papers, they found temporary and then recover. 01:15:18.980 --> 01:15:20.730 align:middle line:90% Some they found persistent. 01:15:20.730 --> 01:15:23.118 align:middle line:84% So then you cannot say you are wrong, you are right-- no, 01:15:23.118 --> 01:15:25.160 align:middle line:84% because they are looking at different hurricanes, 01:15:25.160 --> 01:15:27.830 align:middle line:90% the different data sets. 01:15:27.830 --> 01:15:32.270 align:middle line:84% So algebra-- empirical specification-- 01:15:32.270 --> 01:15:33.000 align:middle line:90% DID. 01:15:33.000 --> 01:15:36.530 align:middle line:90% DID here is log p. 01:15:36.530 --> 01:15:39.990 align:middle line:84% And this is variable name is very complicated. 01:15:39.990 --> 01:15:42.530 align:middle line:90% You just look at this dummy. 01:15:42.530 --> 01:15:47.090 align:middle line:84% Treatment sales in the flood zone times post-- 01:15:47.090 --> 01:15:50.150 align:middle line:84% so here, maybe I need a little bit-- 01:15:50.150 --> 01:15:54.860 align:middle line:84% one-minute explanation of this because otherwise, 01:15:54.860 --> 01:15:56.990 align:middle line:84% when you look at paper, you got confused. 01:15:56.990 --> 01:16:00.050 align:middle line:90% So the point is the following. 01:16:00.050 --> 01:16:03.320 align:middle line:90% You have log p. 01:16:03.320 --> 01:16:05.090 align:middle line:90% That's very clear-- hedonic. 01:16:05.090 --> 01:16:09.780 align:middle line:84% Then you have beta 1 [INAUDIBLE], all the controls, 01:16:09.780 --> 01:16:14.330 align:middle line:84% like number of rooms, like this school, subway. 01:16:14.330 --> 01:16:21.200 align:middle line:84% Then you have another beta times Treat-- 01:16:21.200 --> 01:16:23.070 align:middle line:90% Treat is some dummy of the-- 01:16:23.070 --> 01:16:25.610 align:middle line:90% 01:16:25.610 --> 01:16:27.730 align:middle line:90% yes, OK. 01:16:27.730 --> 01:16:38.720 align:middle line:84% Let's say alpha, beta 1 plus beta 2 times After plus beta 3. 01:16:38.720 --> 01:16:40.500 align:middle line:90% This is the interaction. 01:16:40.500 --> 01:16:45.230 align:middle line:90% 01:16:45.230 --> 01:16:48.180 align:middle line:90% So that's a specification. 01:16:48.180 --> 01:16:50.350 align:middle line:90% So what do you mean by this? 01:16:50.350 --> 01:16:52.740 align:middle line:84% Well, we suddenly become so complicated-- actually, 01:16:52.740 --> 01:16:54.360 align:middle line:90% it's very intuitive. 01:16:54.360 --> 01:16:55.370 align:middle line:90% Think about this. 01:16:55.370 --> 01:16:58.530 align:middle line:90% 01:16:58.530 --> 01:17:00.510 align:middle line:90% Remember this matrix? 01:17:00.510 --> 01:17:05.040 align:middle line:84% This is a treated area, which is a flood zone. 01:17:05.040 --> 01:17:07.890 align:middle line:90% This is a control. 01:17:07.890 --> 01:17:09.280 align:middle line:90% This is before. 01:17:09.280 --> 01:17:10.590 align:middle line:90% This is after. 01:17:10.590 --> 01:17:12.190 align:middle line:90% The force sells. 01:17:12.190 --> 01:17:15.960 align:middle line:84% That's the key thing as we showed here. 01:17:15.960 --> 01:17:18.180 align:middle line:84% Now when you look at the equation-- 01:17:18.180 --> 01:17:24.940 align:middle line:84% when you look at the equation, say the Treat flood zone before, 01:17:24.940 --> 01:17:26.380 align:middle line:90% then Treat is 1-- 01:17:26.380 --> 01:17:28.240 align:middle line:90% Treat is dummy 1. 01:17:28.240 --> 01:17:33.490 align:middle line:84% Then After is 0, and After is 0-- this is 1. 01:17:33.490 --> 01:17:36.610 align:middle line:84% So for this part, we only look at this part. 01:17:36.610 --> 01:17:42.820 align:middle line:84% The only coefficient can survive is beta 1, is that right, before 01:17:42.820 --> 01:17:45.410 align:middle line:90% because before After 0-- 01:17:45.410 --> 01:17:46.310 align:middle line:90% After is the dummy. 01:17:46.310 --> 01:17:47.610 align:middle line:90% Only After is 1. 01:17:47.610 --> 01:17:51.480 align:middle line:90% 0, 0, 0, 0, 0, beta 1. 01:17:51.480 --> 01:17:55.040 align:middle line:90% 01:17:55.040 --> 01:17:58.490 align:middle line:90% And then treat after-- 01:17:58.490 --> 01:17:59.750 align:middle line:90% this survives. 01:17:59.750 --> 01:18:02.090 align:middle line:90% This yes, this yes. 01:18:02.090 --> 01:18:07.880 align:middle line:84% So that's beta 1 plus beta 2 plus beta 3. 01:18:07.880 --> 01:18:11.120 align:middle line:90% Treat after three coefficients. 01:18:11.120 --> 01:18:20.260 align:middle line:90% Control-- before 0, 0, 0, 0. 01:18:20.260 --> 01:18:27.500 align:middle line:90% And After-- 0, 1, 0-- 01:18:27.500 --> 01:18:28.560 align:middle line:90% only beta 2. 01:18:28.560 --> 01:18:32.060 align:middle line:90% 01:18:32.060 --> 01:18:35.530 align:middle line:84% Then you have all the parameters. 01:18:35.530 --> 01:18:42.810 align:middle line:84% And then you do, OK, now I understand that this 01:18:42.810 --> 01:18:47.460 align:middle line:84% difference-in-differences in what kind of thing is 01:18:47.460 --> 01:18:48.430 align:middle line:90% difference-- 01:18:48.430 --> 01:18:57.820 align:middle line:90% 01:18:57.820 --> 01:19:01.520 align:middle line:84% yes, difference, further difference. 01:19:01.520 --> 01:19:06.550 align:middle line:90% 01:19:06.550 --> 01:19:10.270 align:middle line:84% So the thing is this is a measurement. 01:19:10.270 --> 01:19:16.770 align:middle line:84% So when you run this regression, this thing is -8%. 01:19:16.770 --> 01:19:18.720 align:middle line:90% Is that clear? 01:19:18.720 --> 01:19:24.480 align:middle line:84% OK, if you got so confused, come to my table 01:19:24.480 --> 01:19:29.740 align:middle line:84% and think about my apple-to-oranges thing. 01:19:29.740 --> 01:19:36.330 align:middle line:84% This thing then, look, has a coefficient. 01:19:36.330 --> 01:19:39.030 align:middle line:90% Write on your paper you derive-- 01:19:39.030 --> 01:19:42.710 align:middle line:90% this is the coefficient. 01:19:42.710 --> 01:19:45.070 align:middle line:84% AUDIENCE: So what's the thesis of this research? 01:19:45.070 --> 01:19:49.620 align:middle line:84% Is it-- is it that buyers attribute a lower value 01:19:49.620 --> 01:19:52.910 align:middle line:84% to houses that have the potential for flood risk? 01:19:52.910 --> 01:19:53.680 align:middle line:90% SIQI ZHENG: Yes. 01:19:53.680 --> 01:19:56.540 align:middle line:84% AUDIENCE: My issue with that is that this based on what we've 01:19:56.540 --> 01:19:58.610 align:middle line:84% been told, this research seems to ignore 01:19:58.610 --> 01:20:00.830 align:middle line:84% the effects of the increase in insurance premiums 01:20:00.830 --> 01:20:02.640 align:middle line:84% that are the result of post-storm damage. 01:20:02.640 --> 01:20:05.510 align:middle line:84% So you have to go in and rebuild the location, 01:20:05.510 --> 01:20:08.550 align:middle line:84% and then there's going to be a higher holding cost afterwards. 01:20:08.550 --> 01:20:11.090 align:middle line:84% So naturally, there should be some correction there 01:20:11.090 --> 01:20:12.870 align:middle line:84% that will ultimately revert to a mean? 01:20:12.870 --> 01:20:15.925 align:middle line:84% But as far as buyer's preferences, generally, 01:20:15.925 --> 01:20:18.930 align:middle line:84% in the United States, buyers own homes for 5 to 8 years. 01:20:18.930 --> 01:20:20.030 align:middle line:90% So it's not like-- 01:20:20.030 --> 01:20:22.790 align:middle line:84% I don't understand why a rational buyer would say there 01:20:22.790 --> 01:20:27.095 align:middle line:84% was a 1-in-300-year storm that occurred and hit New York. 01:20:27.095 --> 01:20:29.720 align:middle line:84% Yes, there's significant damage, but I project that we're going 01:20:29.720 --> 01:20:31.220 align:middle line:84% to get another 1-in-300-year storm. 01:20:31.220 --> 01:20:34.070 align:middle line:84% Statistically, that's not a rational decision to make. 01:20:34.070 --> 01:20:36.860 align:middle line:84% So I don't think that they can attribute that 01:20:36.860 --> 01:20:38.670 align:middle line:84% to the effect of that massive storm. 01:20:38.670 --> 01:20:41.270 align:middle line:84% It's more the effect of the tail end of it, 01:20:41.270 --> 01:20:44.810 align:middle line:84% which is the insurance industry's response to a large-- 01:20:44.810 --> 01:20:46.410 align:middle line:90% SIQI ZHENG: Yes, yes. 01:20:46.410 --> 01:20:48.540 align:middle line:84% So basically, yes, you are right. 01:20:48.540 --> 01:20:52.920 align:middle line:84% So, for example, if the market is very efficient 01:20:52.920 --> 01:20:55.080 align:middle line:84% and the information-- very efficient, 01:20:55.080 --> 01:20:57.660 align:middle line:84% then even before this happens, they 01:20:57.660 --> 01:21:00.580 align:middle line:84% understand this place is riskier than this place 01:21:00.580 --> 01:21:01.860 align:middle line:90% if you look at the long run. 01:21:01.860 --> 01:21:03.450 align:middle line:84% Then they're already capitalizing 01:21:03.450 --> 01:21:05.080 align:middle line:90% into the place discount. 01:21:05.080 --> 01:21:08.880 align:middle line:84% So after additional shocks-- no change because they fully 01:21:08.880 --> 01:21:12.120 align:middle line:84% understand-- they fully capitalized those future things 01:21:12.120 --> 01:21:12.620 align:middle line:90% in now-- 01:21:12.620 --> 01:21:14.703 align:middle line:84% AUDIENCE: There should be a little bit of a change 01:21:14.703 --> 01:21:16.325 align:middle line:84% though because of the change in-- 01:21:16.325 --> 01:21:17.700 align:middle line:84% SIQI ZHENG: Yeah, temporary, like 01:21:17.700 --> 01:21:20.020 align:middle line:84% you need to repair and things like that. 01:21:20.020 --> 01:21:21.310 align:middle line:90% But later, you will just-- 01:21:21.310 --> 01:21:22.780 align:middle line:90% will become same. 01:21:22.780 --> 01:21:27.100 align:middle line:84% But always, this is not enough information. 01:21:27.100 --> 01:21:29.730 align:middle line:84% So those buyers, at first, they don't have 01:21:29.730 --> 01:21:31.900 align:middle line:90% the big capitalization effect. 01:21:31.900 --> 01:21:34.800 align:middle line:84% Then, after this shock, they understand more. 01:21:34.800 --> 01:21:38.530 align:middle line:84% Then the discount disappear-- discount appear more. 01:21:38.530 --> 01:21:40.630 align:middle line:90% So that's my explanation. 01:21:40.630 --> 01:21:42.610 align:middle line:90% And there's other explanations. 01:21:42.610 --> 01:21:45.720 align:middle line:84% But what you said also makes true because in another paper, 01:21:45.720 --> 01:21:47.610 align:middle line:90% exactly the same data-- 01:21:47.610 --> 01:21:50.100 align:middle line:84% Sandy and the New York City, some of them they found 01:21:50.100 --> 01:21:51.450 align:middle line:90% recovered. 01:21:51.450 --> 01:21:54.420 align:middle line:84% And different neighborhoods are so totally different 01:21:54.420 --> 01:21:56.310 align:middle line:84% in terms of whether they have insurance 01:21:56.310 --> 01:21:58.410 align:middle line:84% and whether they have the capacity to repair 01:21:58.410 --> 01:22:00.120 align:middle line:90% and things like that. 01:22:00.120 --> 01:22:03.700 align:middle line:84% OK, technical thing, any question? 01:22:03.700 --> 01:22:04.896 align:middle line:90% You have a question? 01:22:04.896 --> 01:22:08.370 align:middle line:84% AUDIENCE: [INAUDIBLE] beta 1 before the treat [INAUDIBLE]. 01:22:08.370 --> 01:22:10.890 align:middle line:84% SIQI ZHENG: Beta 1 before, yes, before treat-- 01:22:10.890 --> 01:22:12.390 align:middle line:90% treat is 1. 01:22:12.390 --> 01:22:16.640 align:middle line:84% Treat is always 1 for treated, no matter before or after. 01:22:16.640 --> 01:22:18.730 align:middle line:90% After is 0, and this is 0. 01:22:18.730 --> 01:22:20.520 align:middle line:90% 1 times 0 is 0. 01:22:20.520 --> 01:22:22.840 align:middle line:90% So you only have beta 1 there. 01:22:22.840 --> 01:22:27.860 align:middle line:90% So 1, 0, 1, 0. 01:22:27.860 --> 01:22:31.060 align:middle line:84% Yes, I think this, of course, will take time, too. 01:22:31.060 --> 01:22:32.310 align:middle line:90% At first, I didn't understand. 01:22:32.310 --> 01:22:35.577 align:middle line:84% I need to tell you that when I was a PhD student, 01:22:35.577 --> 01:22:36.410 align:middle line:90% I didn't learn this. 01:22:36.410 --> 01:22:38.990 align:middle line:90% 01:22:38.990 --> 01:22:42.470 align:middle line:84% At that time, they had not-- the econometrics field 01:22:42.470 --> 01:22:44.760 align:middle line:84% advanced very fast in the last 20 years. 01:22:44.760 --> 01:22:48.830 align:middle line:84% When I was a PhD student, no DID, no matching things, no. 01:22:48.830 --> 01:22:50.280 align:middle line:90% And then we didn't learn. 01:22:50.280 --> 01:22:52.620 align:middle line:84% And later, I saw the papers, all of you were using this. 01:22:52.620 --> 01:22:53.960 align:middle line:90% I got so confused. 01:22:53.960 --> 01:22:56.590 align:middle line:84% I didn't understand why this interaction term matters. 01:22:56.590 --> 01:23:00.890 align:middle line:84% I just do all my math, and now I understand. 01:23:00.890 --> 01:23:03.170 align:middle line:90% Good. 01:23:03.170 --> 01:23:06.440 align:middle line:84% Then I will skip this because this I already 01:23:06.440 --> 01:23:09.090 align:middle line:90% told you-- this 8% persistent. 01:23:09.090 --> 01:23:14.330 align:middle line:84% And they also have three level of measures-- 01:23:14.330 --> 01:23:17.330 align:middle line:90% three levels of damage-- 01:23:17.330 --> 01:23:20.220 align:middle line:84% light damage, medium damage, very heavy damage. 01:23:20.220 --> 01:23:22.070 align:middle line:90% You see, it's different. 01:23:22.070 --> 01:23:23.180 align:middle line:90% The heavy damage-- 01:23:23.180 --> 01:23:26.380 align:middle line:90% 15% of the discount. 01:23:26.380 --> 01:23:29.980 align:middle line:84% And these are OK-- overall stabilized at 8%. 01:23:29.980 --> 01:23:33.970 align:middle line:84% For those, first drops a lot and later recovered a little bit 01:23:33.970 --> 01:23:37.030 align:middle line:90% and finally stable at 8%. 01:23:37.030 --> 01:23:38.570 align:middle line:90% So that's for housing. 01:23:38.570 --> 01:23:40.810 align:middle line:84% Remember, housing, housing, housing. 01:23:40.810 --> 01:23:44.720 align:middle line:84% I think, today, I'm not going to have time to for commercial. 01:23:44.720 --> 01:23:47.240 align:middle line:84% But next week, we will talk commercial. 01:23:47.240 --> 01:23:48.820 align:middle line:84% Then there's a commercial real estate 01:23:48.820 --> 01:23:52.700 align:middle line:84% paper says, oh, commercial real estate. 01:23:52.700 --> 01:23:53.510 align:middle line:90% They crave data. 01:23:53.510 --> 01:23:54.340 align:middle line:90% They drop. 01:23:54.340 --> 01:23:56.830 align:middle line:90% After two years-- fully recover. 01:23:56.830 --> 01:23:59.240 align:middle line:84% That's a commercial real estate paper. 01:23:59.240 --> 01:24:01.700 align:middle line:84% But I don't understand why they fully recover. 01:24:01.700 --> 01:24:03.680 align:middle line:84% But anyway, there are so many different papers. 01:24:03.680 --> 01:24:08.340 align:middle line:84% But for this one, 8% in New York City after Sandy. 01:24:08.340 --> 01:24:14.110 align:middle line:84% And also, you see there's other papers. 01:24:14.110 --> 01:24:19.440 align:middle line:84% This is another paper to look at change in the risk premium. 01:24:19.440 --> 01:24:22.890 align:middle line:90% And then you see this lending. 01:24:22.890 --> 01:24:27.600 align:middle line:84% At first, no price differential between flood zone and no flood 01:24:27.600 --> 01:24:28.100 align:middle line:90% zone. 01:24:28.100 --> 01:24:30.290 align:middle line:84% People may be deniers, I don't know. 01:24:30.290 --> 01:24:32.300 align:middle line:84% And they don't have knowledge like that. 01:24:32.300 --> 01:24:36.970 align:middle line:84% And unfortunately, this area got hit several times. 01:24:36.970 --> 01:24:41.070 align:middle line:84% And first, before, no differential, and then one hit. 01:24:41.070 --> 01:24:44.550 align:middle line:90% 01:24:44.550 --> 01:24:48.870 align:middle line:84% After this first hurricane, Fran, got discount. 01:24:48.870 --> 01:24:52.660 align:middle line:84% And then, unfortunately got hit again by the Floyd. 01:24:52.660 --> 01:24:55.830 align:middle line:84% And the further discount-- they learned and learned the-- 01:24:55.830 --> 01:24:59.220 align:middle line:84% to the extent that stable, now they understand, 01:24:59.220 --> 01:25:02.580 align:middle line:84% for this location, the future probability is what, 01:25:02.580 --> 01:25:04.560 align:middle line:90% and they fully capitalize that. 01:25:04.560 --> 01:25:06.760 align:middle line:90% But it still depends on belief. 01:25:06.760 --> 01:25:09.720 align:middle line:84% Maybe no matter how many times, I just don't believe-- 01:25:09.720 --> 01:25:10.680 align:middle line:90% you don't believe. 01:25:10.680 --> 01:25:14.790 align:middle line:90% 01:25:14.790 --> 01:25:18.330 align:middle line:84% AUDIENCE: About this model, even in China, 01:25:18.330 --> 01:25:28.320 align:middle line:84% like one housing [INAUDIBLE] the developer [INAUDIBLE] repair. 01:25:28.320 --> 01:25:32.063 align:middle line:84% And after five years, if the house broken, the-- 01:25:32.063 --> 01:25:33.480 align:middle line:84% SIQI ZHENG: The developer is gone. 01:25:33.480 --> 01:25:37.650 align:middle line:84% AUDIENCE: No, the property [INAUDIBLE] will repair. 01:25:37.650 --> 01:25:41.760 align:middle line:84% But if the climate risk, like flooding, the city government 01:25:41.760 --> 01:25:42.700 align:middle line:90% will repair it. 01:25:42.700 --> 01:25:47.250 align:middle line:84% So the treat and after the [INAUDIBLE] 01:25:47.250 --> 01:25:49.940 align:middle line:84% SIQI ZHENG: Yes, that's something we need to discuss. 01:25:49.940 --> 01:25:52.290 align:middle line:90% Who is responsible to repair? 01:25:52.290 --> 01:25:54.780 align:middle line:84% If always the government responsible to repair, 01:25:54.780 --> 01:25:55.900 align:middle line:90% then they don't care. 01:25:55.900 --> 01:25:58.180 align:middle line:84% They still buy, and the builders still build. 01:25:58.180 --> 01:25:59.980 align:middle line:90% That's a moral hazard. 01:25:59.980 --> 01:26:02.358 align:middle line:84% We call it moral hazard if the thing that-- 01:26:02.358 --> 01:26:03.900 align:middle line:84% you don't bear the risk when you make 01:26:03.900 --> 01:26:06.900 align:middle line:84% the decision, moral hazard-- you just build, build, build. 01:26:06.900 --> 01:26:07.890 align:middle line:90% Yes. 01:26:07.890 --> 01:26:09.400 align:middle line:90% OK, thank you. 01:26:09.400 --> 01:26:11.170 align:middle line:90% And today is about housing. 01:26:11.170 --> 01:26:15.330 align:middle line:84% And because Juan eating my time, I cannot go through this. 01:26:15.330 --> 01:26:18.670 align:middle line:84% Next week, commercial and guest speaker. 01:26:18.670 --> 01:26:20.540 align:middle line:90% Thank you so much. 01:26:20.540 --> 01:26:46.000 align:middle line:90%