WEBVTT 00:00:00.000 --> 00:00:02.000 align:middle line:90% [SQUEAKING] 00:00:02.000 --> 00:00:04.000 align:middle line:90% [RUSTLING] 00:00:04.000 --> 00:00:07.500 align:middle line:90% [CLICKING] 00:00:07.500 --> 00:00:09.342 align:middle line:90% 00:00:09.342 --> 00:00:10.800 align:middle line:84% JON GRUBER: We're going to continue 00:00:10.800 --> 00:00:13.160 align:middle line:90% our discussion of public goods. 00:00:13.160 --> 00:00:15.700 align:middle line:84% You remember the setup in chapter 7? 00:00:15.700 --> 00:00:18.400 align:middle line:84% We talked about general public goods theory 00:00:18.400 --> 00:00:21.960 align:middle line:84% and the general Samuelson rule for optimal public goods 00:00:21.960 --> 00:00:22.960 align:middle line:90% provision. 00:00:22.960 --> 00:00:25.570 align:middle line:84% In chapter 8, we talked about the first aspect 00:00:25.570 --> 00:00:27.320 align:middle line:84% of how you actually deal with the problems 00:00:27.320 --> 00:00:32.960 align:middle line:84% of public good provision, which is cost-benefit analysis. 00:00:32.960 --> 00:00:35.280 align:middle line:84% In chapter 9, we talked about political economy 00:00:35.280 --> 00:00:39.060 align:middle line:84% and all the challenges of accurately creating decisions. 00:00:39.060 --> 00:00:42.360 align:middle line:84% We need to aggregate preferences and get them revealed honestly. 00:00:42.360 --> 00:00:44.080 align:middle line:84% Now in chapter 10 and 11, we're going 00:00:44.080 --> 00:00:47.040 align:middle line:84% to do two applications of public good theory. 00:00:47.040 --> 00:00:50.600 align:middle line:84% Today, we'll talk about what we call fiscal federalism, which 00:00:50.600 --> 00:00:53.860 align:middle line:84% is this question I raised in chapter 1 of what should be done 00:00:53.860 --> 00:00:56.020 align:middle line:90% at which level of government. 00:00:56.020 --> 00:00:58.840 align:middle line:84% Why is, for example, national defense a federal priority 00:00:58.840 --> 00:01:00.955 align:middle line:84% while fixing road is a local priority? 00:01:00.955 --> 00:01:02.580 align:middle line:84% So that's the first thing we'll discuss 00:01:02.580 --> 00:01:04.300 align:middle line:90% is fiscal federalism today. 00:01:04.300 --> 00:01:05.920 align:middle line:84% Next time, and maybe starting today, 00:01:05.920 --> 00:01:07.700 align:middle line:84% we'll start talking about education 00:01:07.700 --> 00:01:10.975 align:middle line:84% as an example of a government provided public good. 00:01:10.975 --> 00:01:12.600 align:middle line:84% But let's start with fiscal federalism. 00:01:12.600 --> 00:01:14.933 align:middle line:84% This is particularly timely, more so than any other time 00:01:14.933 --> 00:01:17.740 align:middle line:84% I taught this class, probably, which 00:01:17.740 --> 00:01:22.118 align:middle line:84% is that in the wake of COVID, we saw an unprecedented transfer 00:01:22.118 --> 00:01:24.160 align:middle line:84% from the federal government to state governments. 00:01:24.160 --> 00:01:25.900 align:middle line:84% If look at all the bills that responded 00:01:25.900 --> 00:01:28.800 align:middle line:84% to COVID, all the fiscal relief bills, in total, 00:01:28.800 --> 00:01:33.220 align:middle line:84% they transferred $350 billion from the federal government 00:01:33.220 --> 00:01:34.860 align:middle line:90% to the state government. 00:01:34.860 --> 00:01:35.880 align:middle line:90% And that's a big amount. 00:01:35.880 --> 00:01:37.660 align:middle line:90% And the question is why? 00:01:37.660 --> 00:01:39.190 align:middle line:84% Why is the federal government-- why 00:01:39.190 --> 00:01:40.920 align:middle line:84% didn't the federal government just spend the money? 00:01:40.920 --> 00:01:42.503 align:middle line:84% Why did they feel the need to transfer 00:01:42.503 --> 00:01:43.940 align:middle line:90% to state and local governments? 00:01:43.940 --> 00:01:47.020 align:middle line:84% And that's exactly the kind of question we'll answer today. 00:01:47.020 --> 00:01:51.980 align:middle line:84% So to look at that, let's start with some facts. 00:01:51.980 --> 00:01:54.280 align:middle line:84% The first fact is in figure 10-1. 00:01:54.280 --> 00:01:57.540 align:middle line:84% So the first handout is for chapter 10. 00:01:57.540 --> 00:02:02.760 align:middle line:84% Figure 10-1 is the green, purple, and blue bars. 00:02:02.760 --> 00:02:06.960 align:middle line:84% This shows the division of spending across different levels 00:02:06.960 --> 00:02:09.360 align:middle line:90% of government over time. 00:02:09.360 --> 00:02:15.400 align:middle line:84% So in 1902, more than a century ago, 122 years ago, 00:02:15.400 --> 00:02:19.200 align:middle line:84% most spending in the US, that's government spending, 00:02:19.200 --> 00:02:21.680 align:middle line:90% was done by local governments. 00:02:21.680 --> 00:02:23.933 align:middle line:84% There was a very small role for states 00:02:23.933 --> 00:02:26.100 align:middle line:84% and a fairly modest role for the federal government. 00:02:26.100 --> 00:02:29.120 align:middle line:84% Indeed, that was true through the 1920s. 00:02:29.120 --> 00:02:32.820 align:middle line:84% So by the time we get to the Great Depression in 1927, 00:02:32.820 --> 00:02:35.120 align:middle line:84% most government activity happens at the level 00:02:35.120 --> 00:02:37.120 align:middle line:90% of local governments. 00:02:37.120 --> 00:02:42.880 align:middle line:84% That changes dramatically between 1927 and 1952, 00:02:42.880 --> 00:02:46.720 align:middle line:84% where the centralized share spending goes up dramatically 00:02:46.720 --> 00:02:48.500 align:middle line:90% and the local share shrinks. 00:02:48.500 --> 00:02:50.640 align:middle line:84% States there stays about the same. 00:02:50.640 --> 00:02:51.420 align:middle line:90% What changed? 00:02:51.420 --> 00:02:53.360 align:middle line:90% Well, two things changed. 00:02:53.360 --> 00:02:58.050 align:middle line:84% First of all was World War II, which dramatically increased 00:02:58.050 --> 00:03:02.330 align:middle line:84% defense spending, which is done at the central government level. 00:03:02.330 --> 00:03:05.330 align:middle line:84% The second was the New Deal, which 00:03:05.330 --> 00:03:08.650 align:middle line:84% is a broad set of new federal programs 00:03:08.650 --> 00:03:12.810 align:middle line:84% to support our nation's poor and elderly. 00:03:12.810 --> 00:03:16.850 align:middle line:84% By 1977, and especially by 2002, you 00:03:16.850 --> 00:03:19.570 align:middle line:84% see an interesting trend, which is the federal government 00:03:19.570 --> 00:03:24.450 align:middle line:84% share shrink, but much of it's made up for by the state share. 00:03:24.450 --> 00:03:28.610 align:middle line:84% So you see over this period, you see a shift. 00:03:28.610 --> 00:03:30.953 align:middle line:84% Local government stays roughly the same size. 00:03:30.953 --> 00:03:33.370 align:middle line:84% You see a shift of less federal responsibility, more state 00:03:33.370 --> 00:03:34.930 align:middle line:90% responsibility. 00:03:34.930 --> 00:03:38.250 align:middle line:84% And in particular, this period saw a large rise in what we 00:03:38.250 --> 00:03:41.210 align:middle line:90% call intergovernmental grants. 00:03:41.210 --> 00:03:44.410 align:middle line:84% Intergovernmental grants are transfers from one level 00:03:44.410 --> 00:03:46.210 align:middle line:90% of government to the other. 00:03:46.210 --> 00:03:53.570 align:middle line:84% And so this period from '77 to 2002 and beyond 00:03:53.570 --> 00:03:56.390 align:middle line:84% or really from '52 to 2002 and beyond 00:03:56.390 --> 00:03:59.790 align:middle line:84% saw a large rise in intergovernmental grants. 00:03:59.790 --> 00:04:03.070 align:middle line:84% So the first half of the 20th century was marked by a rise 00:04:03.070 --> 00:04:04.657 align:middle line:90% in the central government. 00:04:04.657 --> 00:04:06.990 align:middle line:84% The second half of the 20th century was marked by a rise 00:04:06.990 --> 00:04:09.350 align:middle line:84% in state government, [? inside of ?] state government, 00:04:09.350 --> 00:04:11.790 align:middle line:84% but not necessarily funded by state governments. 00:04:11.790 --> 00:04:15.430 align:middle line:84% It was a transfer of the federal government offsetting 00:04:15.430 --> 00:04:18.649 align:middle line:84% some of the spending it does to states. 00:04:18.649 --> 00:04:20.510 align:middle line:84% And this, in particular, came through things 00:04:20.510 --> 00:04:22.830 align:middle line:84% like the Medicaid program, which is 00:04:22.830 --> 00:04:27.790 align:middle line:84% a jointly funded state-federal health insurance program. 00:04:27.790 --> 00:04:29.730 align:middle line:84% So there was more joint funding with states. 00:04:29.730 --> 00:04:31.278 align:middle line:84% At the same time, there was more-- 00:04:31.278 --> 00:04:33.070 align:middle line:84% basically, the federal government basically 00:04:33.070 --> 00:04:36.270 align:middle line:84% gives states more responsibility and sent them more money 00:04:36.270 --> 00:04:41.070 align:middle line:84% to be able to fund those responsibilities. 00:04:41.070 --> 00:04:44.230 align:middle line:84% And then things are pretty things are pretty stabilized. 00:04:44.230 --> 00:04:49.230 align:middle line:84% Now, what do states and localities spend their money on? 00:04:49.230 --> 00:04:51.050 align:middle line:84% We talked about that in chapter 1. 00:04:51.050 --> 00:04:54.210 align:middle line:84% They spend their money on health just like the federal government 00:04:54.210 --> 00:04:57.930 align:middle line:84% does, on a modest amount of redistribution 00:04:57.930 --> 00:05:00.090 align:middle line:84% in social insurance, but a lot of things 00:05:00.090 --> 00:05:02.930 align:middle line:84% like schools and roads and police. 00:05:02.930 --> 00:05:09.472 align:middle line:84% Now, what's interesting is if you look at across states, 00:05:09.472 --> 00:05:11.930 align:middle line:84% two main sources of spending are education and health care. 00:05:11.930 --> 00:05:15.250 align:middle line:84% And table 1 shows how much variation 00:05:15.250 --> 00:05:17.010 align:middle line:90% there is across states. 00:05:17.010 --> 00:05:20.090 align:middle line:84% There's huge variation in what states spend on things 00:05:20.090 --> 00:05:23.970 align:middle line:90% like education and health care. 00:05:23.970 --> 00:05:27.290 align:middle line:84% There's also, by the way, huge variation in what states raise. 00:05:27.290 --> 00:05:30.690 align:middle line:84% For instance, New York collects almost $3,000 per person 00:05:30.690 --> 00:05:36.290 align:middle line:84% in income taxes, whereas seven states have no income tax. 00:05:36.290 --> 00:05:39.570 align:middle line:84% In terms of sales taxes, Washington raised $2,500 per 00:05:39.570 --> 00:05:42.570 align:middle line:84% person per year in sales taxes, while four states literally have 00:05:42.570 --> 00:05:44.650 align:middle line:90% no sales tax. 00:05:44.650 --> 00:05:47.890 align:middle line:84% So you've got a situation where there's enormous differences-- 00:05:47.890 --> 00:05:51.550 align:middle line:84% the main takeaway from table 10-1 is that there are enormous 00:05:51.550 --> 00:05:54.350 align:middle line:84% differences across states in how they spend their money. 00:05:54.350 --> 00:05:58.230 align:middle line:84% So when the federal government devolves responsibility 00:05:58.230 --> 00:06:01.850 align:middle line:84% to the states, it's increasing the variance in how 00:06:01.850 --> 00:06:03.218 align:middle line:90% its citizens are treated. 00:06:03.218 --> 00:06:05.010 align:middle line:84% When the federal government does something, 00:06:05.010 --> 00:06:07.397 align:middle line:84% its citizens are treated fairly, uniformly nationally. 00:06:07.397 --> 00:06:09.730 align:middle line:84% When the federal government devolves that responsibility 00:06:09.730 --> 00:06:11.530 align:middle line:84% to states and then even more to localities, 00:06:11.530 --> 00:06:13.928 align:middle line:84% that raises the variance in how citizens are treated. 00:06:13.928 --> 00:06:15.470 align:middle line:84% And that's an interesting observation 00:06:15.470 --> 00:06:17.670 align:middle line:90% want to carry with us today. 00:06:17.670 --> 00:06:22.030 align:middle line:84% Now, interestingly, we are much more decentralized 00:06:22.030 --> 00:06:22.970 align:middle line:90% than most nations. 00:06:22.970 --> 00:06:31.190 align:middle line:84% If you look at table 10-2, this shows the share of spending 00:06:31.190 --> 00:06:34.385 align:middle line:84% and revenues that are done by subnational governments, 00:06:34.385 --> 00:06:36.510 align:middle line:84% what we call state and local governments in the US. 00:06:36.510 --> 00:06:39.730 align:middle line:84% Basically, non-centralized share spending and revenue. 00:06:39.730 --> 00:06:43.610 align:middle line:84% So what you see is in the US, about 43% of-- it's 00:06:43.610 --> 00:06:44.990 align:middle line:90% the third from the bottom row. 00:06:44.990 --> 00:06:48.790 align:middle line:84% About 43% of our spending and 48% of our revenues 00:06:48.790 --> 00:06:51.850 align:middle line:84% are collected at the state level. 00:06:51.850 --> 00:06:53.850 align:middle line:84% That is much, much higher than the average 00:06:53.850 --> 00:06:55.850 align:middle line:90% the rest of the world. 00:06:55.850 --> 00:06:58.710 align:middle line:84% The OECD average, developed country, 00:06:58.710 --> 00:07:02.670 align:middle line:84% they spend 31% at the subnational level, 00:07:02.670 --> 00:07:03.890 align:middle line:90% where we spend 43%. 00:07:03.890 --> 00:07:06.950 align:middle line:84% And they collect 19% at the subnational level, 00:07:06.950 --> 00:07:09.690 align:middle line:84% whereas we collect almost 50% of our revenues. 00:07:09.690 --> 00:07:12.410 align:middle line:84% So we are much more decentralized 00:07:12.410 --> 00:07:14.870 align:middle line:84% than most developed countries, not all. 00:07:14.870 --> 00:07:16.930 align:middle line:84% You can see Denmark's even more decentralized 00:07:16.930 --> 00:07:18.410 align:middle line:90% on the spending side. 00:07:18.410 --> 00:07:20.250 align:middle line:84% But we're much more decentralized 00:07:20.250 --> 00:07:23.130 align:middle line:84% than most, although, interestingly, the trend 00:07:23.130 --> 00:07:26.578 align:middle line:84% worldwide has been towards more decentralization over time. 00:07:26.578 --> 00:07:28.370 align:middle line:84% There's still more centralized, but there's 00:07:28.370 --> 00:07:31.930 align:middle line:84% been a trend towards more decentralization. 00:07:31.930 --> 00:07:34.210 align:middle line:84% Now, this raises the question-- the fact 00:07:34.210 --> 00:07:36.370 align:middle line:84% that there's so much variation raise 00:07:36.370 --> 00:07:39.730 align:middle line:84% the question of what is the right differentiation 00:07:39.730 --> 00:07:42.870 align:middle line:84% of responsibilities across different levels of government 00:07:42.870 --> 00:07:46.290 align:middle line:84% or what we call optimal fiscal federalism? 00:07:46.290 --> 00:07:49.500 align:middle line:84% Fiscal federalism is the term for the distribution 00:07:49.500 --> 00:07:53.860 align:middle line:84% of responsibilities across different levels of government. 00:07:53.860 --> 00:07:55.900 align:middle line:84% Now, what we're going to start with 00:07:55.900 --> 00:08:01.200 align:middle line:84% is the workhorse economics model for optimal fiscal federalism, 00:08:01.200 --> 00:08:03.900 align:middle line:84% what's known as the Tiebout model. 00:08:03.900 --> 00:08:06.040 align:middle line:84% It looks like "tbout," but it's pronounced "tebow" 00:08:06.040 --> 00:08:08.580 align:middle line:84% like the old quarterback for the sports fans among us, 00:08:08.580 --> 00:08:10.900 align:middle line:90% the Tiebout model. 00:08:10.900 --> 00:08:13.700 align:middle line:84% And basically, Charles Tiebout was 00:08:13.700 --> 00:08:16.660 align:middle line:84% an economist who in 1956 said, look, 00:08:16.660 --> 00:08:18.920 align:middle line:84% the Samuelson article had come out. 00:08:18.920 --> 00:08:21.780 align:middle line:84% We all knew that there was too little provision of public goods 00:08:21.780 --> 00:08:23.160 align:middle line:90% compared to private goods. 00:08:23.160 --> 00:08:24.868 align:middle line:84% And Tiebout asked the following question. 00:08:24.868 --> 00:08:29.100 align:middle line:84% He said, what is it that makes private good provision 00:08:29.100 --> 00:08:32.260 align:middle line:84% efficient, where public good provision by the private market 00:08:32.260 --> 00:08:33.620 align:middle line:90% is not? 00:08:33.620 --> 00:08:35.140 align:middle line:90% What's the key? 00:08:35.140 --> 00:08:39.220 align:middle line:84% And he said, the key thing missing for public goods 00:08:39.220 --> 00:08:42.780 align:middle line:84% is essentially shopping and competition 00:08:42.780 --> 00:08:45.940 align:middle line:84% And the market forces that drive the private market 00:08:45.940 --> 00:08:46.985 align:middle line:90% towards efficiency. 00:08:46.985 --> 00:08:48.860 align:middle line:84% What drives the private market to efficiency? 00:08:48.860 --> 00:08:50.480 align:middle line:84% Why is the private market efficient? 00:08:50.480 --> 00:08:52.580 align:middle line:84% Because there's shopping and competition. 00:08:52.580 --> 00:08:54.455 align:middle line:84% When there's imperfect competition, remember, 00:08:54.455 --> 00:08:56.240 align:middle line:84% the private market is no longer efficient. 00:08:56.240 --> 00:08:57.990 align:middle line:84% So what makes the private market efficient 00:08:57.990 --> 00:08:59.640 align:middle line:90% is shopping competition. 00:08:59.640 --> 00:09:04.140 align:middle line:84% And he said, look, that's just not true for public goods. 00:09:04.140 --> 00:09:05.900 align:middle line:84% You don't shop across defense systems. 00:09:05.900 --> 00:09:08.840 align:middle line:84% I mean, you could move across countries to the one we like 00:09:08.840 --> 00:09:11.133 align:middle line:84% the national defense the most, but you probably won't. 00:09:11.133 --> 00:09:13.800 align:middle line:84% It's a pretty modest compared to culture and language and family 00:09:13.800 --> 00:09:14.720 align:middle line:90% ties. 00:09:14.720 --> 00:09:18.960 align:middle line:84% So there's not a lot of shopping across public goods 00:09:18.960 --> 00:09:20.720 align:middle line:90% at the federal level. 00:09:20.720 --> 00:09:25.720 align:middle line:84% But he said at locality level, there is. 00:09:25.720 --> 00:09:29.560 align:middle line:84% In fact, at the locality level, people very much do shop. 00:09:29.560 --> 00:09:32.080 align:middle line:84% They vote with their feet, which is 00:09:32.080 --> 00:09:37.320 align:middle line:84% by moving to localities that can be otherwise similar, 00:09:37.320 --> 00:09:39.800 align:middle line:84% but which can be more closely matched 00:09:39.800 --> 00:09:42.360 align:middle line:90% to their taste for public goods. 00:09:42.360 --> 00:09:48.140 align:middle line:84% And indeed, he argued, that in the limit, 00:09:48.140 --> 00:09:51.260 align:middle line:84% you can actually get efficient public good provision 00:09:51.260 --> 00:09:55.280 align:middle line:84% in the private market through shopping across localities. 00:09:55.280 --> 00:09:58.940 align:middle line:84% It's a coast-like result, which is that basically you 00:09:58.940 --> 00:10:00.900 align:middle line:84% can actually solve the Samuelson public goods 00:10:00.900 --> 00:10:05.220 align:middle line:84% problem by letting people shop for public goods 00:10:05.220 --> 00:10:06.600 align:middle line:90% across localities. 00:10:06.600 --> 00:10:10.460 align:middle line:90% 00:10:10.460 --> 00:10:18.697 align:middle line:84% So his point is, look, let's say you found that in the 1980s, 00:10:18.697 --> 00:10:20.780 align:middle line:84% we were told that the US Department of Defense was 00:10:20.780 --> 00:10:24.660 align:middle line:84% spending, on average, $110 on electric diodes that were worth 00:10:24.660 --> 00:10:30.787 align:middle line:84% $0.04 or $435 on claw hammers or $437 on tape measures. 00:10:30.787 --> 00:10:33.120 align:middle line:84% And we all got outraged and said, what a waste of money, 00:10:33.120 --> 00:10:34.860 align:middle line:90% but no one moved over it. 00:10:34.860 --> 00:10:38.820 align:middle line:84% On the other hand, in New York City in 2014, 00:10:38.820 --> 00:10:41.180 align:middle line:84% they showed that the public library system was spending 00:10:41.180 --> 00:10:44.920 align:middle line:84% $37,000 a year to provide the library director with a sports 00:10:44.920 --> 00:10:46.807 align:middle line:84% car and raising taxes accordingly. 00:10:46.807 --> 00:10:49.140 align:middle line:84% At that point, you might say, look, I'm out of New York. 00:10:49.140 --> 00:10:49.940 align:middle line:90% It's got high taxes. 00:10:49.940 --> 00:10:51.560 align:middle line:84% I don't really care if the library director has a sports 00:10:51.560 --> 00:10:52.120 align:middle line:90% car. 00:10:52.120 --> 00:10:53.260 align:middle line:90% I'll just move to the suburbs. 00:10:53.260 --> 00:10:54.343 align:middle line:90% I can still get to my job. 00:10:54.343 --> 00:10:55.560 align:middle line:90% What do I care? 00:10:55.560 --> 00:10:57.220 align:middle line:90% So you can vote with your feet. 00:10:57.220 --> 00:11:01.640 align:middle line:84% And to boost that on a model under which you can actually 00:11:01.640 --> 00:11:04.280 align:middle line:84% solve the Samuelson problem through competition 00:11:04.280 --> 00:11:07.800 align:middle line:84% across localities-- and the model is quite simple. 00:11:07.800 --> 00:11:10.680 align:middle line:84% Imagine that there are many people in the world, 00:11:10.680 --> 00:11:17.940 align:middle line:84% and they divide themselves into N towns with N sub i residents. 00:11:17.940 --> 00:11:21.240 align:middle line:90% 00:11:21.240 --> 00:11:23.185 align:middle line:90% There are N towns. 00:11:23.185 --> 00:11:24.060 align:middle line:90% I'm sorry, I'm sorry. 00:11:24.060 --> 00:11:25.512 align:middle line:90% No, that's backwards. 00:11:25.512 --> 00:11:26.220 align:middle line:90% That's backwards. 00:11:26.220 --> 00:11:27.800 align:middle line:90% Scratch what I said. 00:11:27.800 --> 00:11:30.220 align:middle line:90% There are towns denoted by i. 00:11:30.220 --> 00:11:31.720 align:middle line:90% There's towns denoted by i. 00:11:31.720 --> 00:11:33.635 align:middle line:90% Each town has N sub i residents. 00:11:33.635 --> 00:11:34.760 align:middle line:90% That's what I meant to say. 00:11:34.760 --> 00:11:36.400 align:middle line:90% They're towns denoted by i. 00:11:36.400 --> 00:11:39.920 align:middle line:84% And each town has N sub i residents. 00:11:39.920 --> 00:11:44.360 align:middle line:84% And let's say moreover that towns finance any public good 00:11:44.360 --> 00:11:46.980 align:middle line:84% they're going to provide with a flat tax 00:11:46.980 --> 00:11:50.640 align:middle line:84% G sub i, which is equal for all i. 00:11:50.640 --> 00:11:53.620 align:middle line:90% It's a flat lump sum tax. 00:11:53.620 --> 00:11:56.220 align:middle line:84% So each town i finances public goods 00:11:56.220 --> 00:11:58.940 align:middle line:84% with a flat, what we call a lump sum, a lump sum 00:11:58.940 --> 00:12:01.260 align:middle line:90% tax, a flat amount on everyone. 00:12:01.260 --> 00:12:03.200 align:middle line:84% Doesn't depend on anything about you. 00:12:03.200 --> 00:12:04.640 align:middle line:90% It's not individualistic. 00:12:04.640 --> 00:12:08.420 align:middle line:84% It's just a flat tax on everyone that's equal. 00:12:08.420 --> 00:12:15.580 align:middle line:84% So there's N sub i residents in the town, and they each pay. 00:12:15.580 --> 00:12:19.120 align:middle line:90% And there's a flat tax. 00:12:19.120 --> 00:12:21.140 align:middle line:84% The total public goods version is G sub i. 00:12:21.140 --> 00:12:21.740 align:middle line:90% I'm sorry. 00:12:21.740 --> 00:12:23.780 align:middle line:84% The total public goods price G sub i. 00:12:23.780 --> 00:12:29.100 align:middle line:84% So each resident pays G sub i over N sub i in taxes. 00:12:29.100 --> 00:12:33.880 align:middle line:84% N sub i residents, G sub i is the total level of public goods. 00:12:33.880 --> 00:12:37.940 align:middle line:84% So each resident pays a flat tax of G sub i over N sub i. 00:12:37.940 --> 00:12:40.160 align:middle line:84% That is the way the system works, 00:12:40.160 --> 00:12:42.790 align:middle line:90% i towns, N sub i people in each. 00:12:42.790 --> 00:12:44.837 align:middle line:84% G sub i is what they spend on public goods. 00:12:44.837 --> 00:12:45.670 align:middle line:90% It can be different. 00:12:45.670 --> 00:12:46.710 align:middle line:90% Each i can be different. 00:12:46.710 --> 00:12:50.630 align:middle line:84% And they charge G sub i over N sub i for public goods. 00:12:50.630 --> 00:12:54.750 align:middle line:84% He argues that in that world we will actually 00:12:54.750 --> 00:12:58.070 align:middle line:84% get efficient public goods provision solely 00:12:58.070 --> 00:13:00.230 align:middle line:90% from voting with your feet. 00:13:00.230 --> 00:13:02.550 align:middle line:84% And the reason is because in this world, 00:13:02.550 --> 00:13:04.833 align:middle line:84% we can get back to the Lindahl equilibrium. 00:13:04.833 --> 00:13:06.750 align:middle line:84% Remember, the problem with Lindahl equilibrium 00:13:06.750 --> 00:13:08.950 align:middle line:84% is people had to tell us the truth. 00:13:08.950 --> 00:13:11.270 align:middle line:84% Well, in this world, people reveal the truth 00:13:11.270 --> 00:13:16.350 align:middle line:84% by moving to the places that reveal their true preferences. 00:13:16.350 --> 00:13:18.530 align:middle line:84% They move to what reveals their true preferences. 00:13:18.530 --> 00:13:21.110 align:middle line:84% Therefore, we get people getting exactly the amount 00:13:21.110 --> 00:13:24.067 align:middle line:84% of public goods they want through private behavior. 00:13:24.067 --> 00:13:26.150 align:middle line:84% Let's go to the example we talked about in chapter 00:13:26.150 --> 00:13:27.670 align:middle line:90% 7 and chapter 9. 00:13:27.670 --> 00:13:29.750 align:middle line:84% Remember Jack and Ava and the fireworks? 00:13:29.750 --> 00:13:38.590 align:middle line:84% You remember Jack wanted 75 fireworks at $0.75 each. 00:13:38.590 --> 00:13:44.650 align:middle line:84% Ava wanted 25 fireworks at $0.25 each. 00:13:44.650 --> 00:13:47.130 align:middle line:84% The Samuelson rule was we should end up 00:13:47.130 --> 00:13:51.970 align:middle line:90% providing 100 fireworks at $1. 00:13:51.970 --> 00:13:54.570 align:middle line:90% That was the Samuelson rule. 00:13:54.570 --> 00:13:57.450 align:middle line:84% But we then talked about how Jack and Ava won't necessarily 00:13:57.450 --> 00:14:01.010 align:middle line:84% get there because Jack has an incentive to lie, 00:14:01.010 --> 00:14:05.290 align:middle line:84% understate his preferences, and by lying, he makes Ava pay more. 00:14:05.290 --> 00:14:08.350 align:middle line:84% And so he pays less but gets almost as many fireworks. 00:14:08.350 --> 00:14:09.170 align:middle line:90% Steven. 00:14:09.170 --> 00:14:11.090 align:middle line:84% AUDIENCE: Does the example [INAUDIBLE] 00:14:11.090 --> 00:14:14.475 align:middle line:84% account for a representative system? 00:14:14.475 --> 00:14:15.850 align:middle line:84% JON GRUBER: Right now, we're not. 00:14:15.850 --> 00:14:18.100 align:middle line:84% We're going to come back to limitations of this model. 00:14:18.100 --> 00:14:20.650 align:middle line:84% But right now, just assume everybody just kind of-- there's 00:14:20.650 --> 00:14:21.350 align:middle line:90% one. 00:14:21.350 --> 00:14:22.150 align:middle line:90% The people vote. 00:14:22.150 --> 00:14:22.872 align:middle line:90% They all decide. 00:14:22.872 --> 00:14:24.830 align:middle line:84% Well, actually, this is the point of the proof. 00:14:24.830 --> 00:14:26.090 align:middle line:84% So basically, Jack and Ava are that. 00:14:26.090 --> 00:14:27.070 align:middle line:90% We said they cheat. 00:14:27.070 --> 00:14:31.490 align:middle line:84% Now let's say Jack and Ava should have their own towns. 00:14:31.490 --> 00:14:34.382 align:middle line:84% So in chapter 9, we said Jack had incentive 00:14:34.382 --> 00:14:35.590 align:middle line:90% to lie about his preferences. 00:14:35.590 --> 00:14:37.770 align:middle line:84% We proved that, that he would get a better deal. 00:14:37.770 --> 00:14:41.290 align:middle line:84% But now let's say Jack and Ava have their own towns. 00:14:41.290 --> 00:14:45.830 align:middle line:84% And so the only way Jack can lie is by moving to Ava's town. 00:14:45.830 --> 00:14:47.538 align:middle line:84% What happens if Jack moves to Ava's town? 00:14:47.538 --> 00:14:49.122 align:middle line:84% Well, he gets a third of the fireworks 00:14:49.122 --> 00:14:50.250 align:middle line:90% at a third of the cost. 00:14:50.250 --> 00:14:53.230 align:middle line:90% He's no better off. 00:14:53.230 --> 00:14:55.090 align:middle line:84% Before he was better off because by voting, 00:14:55.090 --> 00:14:57.470 align:middle line:84% he could bring Ava into the pool and make her pay more. 00:14:57.470 --> 00:14:59.810 align:middle line:84% Now he can't because there's a town full of Avas, 00:14:59.810 --> 00:15:02.790 align:middle line:84% he's got to move there, he's got to do what she wants. 00:15:02.790 --> 00:15:05.350 align:middle line:84% By voting with his feet, we reveal 00:15:05.350 --> 00:15:07.330 align:middle line:84% that Jack will stay at what's optimal for him. 00:15:07.330 --> 00:15:09.990 align:middle line:84% There'll be a set of towns of people with different fireworks 00:15:09.990 --> 00:15:13.438 align:middle line:84% preferences, and everyone will be in exactly the town that 00:15:13.438 --> 00:15:15.230 align:middle line:84% matches their fireworks preferences, paying 00:15:15.230 --> 00:15:17.238 align:middle line:90% exactly the right price. 00:15:17.238 --> 00:15:19.530 align:middle line:84% And the reason is because there's no incentive to move. 00:15:19.530 --> 00:15:23.738 align:middle line:84% You're in a Nash equilibrium because every other town, 00:15:23.738 --> 00:15:25.530 align:middle line:84% if you get a different amount of fireworks, 00:15:25.530 --> 00:15:27.613 align:middle line:84% you pay the right amount for that different amount 00:15:27.613 --> 00:15:28.790 align:middle line:90% of fireworks. 00:15:28.790 --> 00:15:30.453 align:middle line:84% So by differentiating these towns, when 00:15:30.453 --> 00:15:31.870 align:middle line:84% people vote with their feet, we've 00:15:31.870 --> 00:15:33.530 align:middle line:84% achieved the Lindahl equilibrium. 00:15:33.530 --> 00:15:34.890 align:middle line:90% We're not relying on honesty. 00:15:34.890 --> 00:15:37.530 align:middle line:90% We're relying on people moving. 00:15:37.530 --> 00:15:39.790 align:middle line:84% And that is essentially-- the formal proof. 00:15:39.790 --> 00:15:43.130 align:middle line:84% But that's essentially the Tiebout argument, 00:15:43.130 --> 00:15:46.303 align:middle line:84% is that basically local public goods 00:15:46.303 --> 00:15:48.470 align:middle line:84% will be provided efficiently because if they're not, 00:15:48.470 --> 00:15:49.575 align:middle line:90% people will just move. 00:15:49.575 --> 00:15:50.950 align:middle line:84% That's the way to think about it. 00:15:50.950 --> 00:15:52.630 align:middle line:84% If you're ever in a town you don't like the public goods, 00:15:52.630 --> 00:15:54.010 align:middle line:90% you move to one where you do. 00:15:54.010 --> 00:15:56.750 align:middle line:84% And therefore, everyone will get the public goods they want. 00:15:56.750 --> 00:15:57.250 align:middle line:90% OK, yeah? 00:15:57.250 --> 00:15:59.042 align:middle line:84% AUDIENCE: Is that what you mean by shopping 00:15:59.042 --> 00:16:02.630 align:middle line:84% with your feet [INAUDIBLE] all together in town? 00:16:02.630 --> 00:16:05.030 align:middle line:84% JON GRUBER: Yeah, they're moving into all these towns. 00:16:05.030 --> 00:16:07.090 align:middle line:84% There's these i towns that have N people. 00:16:07.090 --> 00:16:09.370 align:middle line:84% In each town, they have G level of public goods. 00:16:09.370 --> 00:16:10.770 align:middle line:90% And you move across towns. 00:16:10.770 --> 00:16:11.830 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:16:11.830 --> 00:16:15.490 align:middle line:84% JON GRUBER: Yeah, you move across towns 00:16:15.490 --> 00:16:17.570 align:middle line:84% until you get to the town that you like. 00:16:17.570 --> 00:16:18.270 align:middle line:90% Good question. 00:16:18.270 --> 00:16:18.890 align:middle line:90% Sorry. 00:16:18.890 --> 00:16:19.430 align:middle line:90% Yep. 00:16:19.430 --> 00:16:21.210 align:middle line:84% It's a colloquialism that's in my head. 00:16:21.210 --> 00:16:23.887 align:middle line:84% But you're right, it's not obvious until I explain it. 00:16:23.887 --> 00:16:24.470 align:middle line:90% Good question. 00:16:24.470 --> 00:16:27.130 align:middle line:90% Other questions? 00:16:27.130 --> 00:16:29.330 align:middle line:90% So that's the Tiebout proof. 00:16:29.330 --> 00:16:32.490 align:middle line:84% And once again, like [? Cos, ?] like the other things 00:16:32.490 --> 00:16:35.970 align:middle line:84% we've talked about, it's a very interesting tuition, 00:16:35.970 --> 00:16:38.012 align:middle line:84% which is clearly not true in practice. 00:16:38.012 --> 00:16:39.470 align:middle line:84% There's clearly a number of reasons 00:16:39.470 --> 00:16:41.930 align:middle line:84% why this won't work perfectly in practice. 00:16:41.930 --> 00:16:43.790 align:middle line:90% And what are those? 00:16:43.790 --> 00:16:46.990 align:middle line:84% Basically, in subsets, there's two classes of reasons. 00:16:46.990 --> 00:16:50.550 align:middle line:84% The first is reasons why this mechanism of competition 00:16:50.550 --> 00:16:51.978 align:middle line:90% won't actually work well. 00:16:51.978 --> 00:16:54.270 align:middle line:84% So what are some reasons why this competitive mechanism 00:16:54.270 --> 00:16:55.370 align:middle line:90% won't actually work well? 00:16:55.370 --> 00:16:55.870 align:middle line:90% Yeah? 00:16:55.870 --> 00:16:58.062 align:middle line:84% AUDIENCE: Moving different town [INAUDIBLE]. 00:16:58.062 --> 00:16:58.770 align:middle line:90% JON GRUBER: Yeah. 00:16:58.770 --> 00:17:01.090 align:middle line:84% So basically, you've got costly moving. 00:17:01.090 --> 00:17:03.822 align:middle line:90% 00:17:03.822 --> 00:17:06.030 align:middle line:84% So you're not going to move until there's some hurdle 00:17:06.030 --> 00:17:07.270 align:middle line:84% rate in which you're really pissed off 00:17:07.270 --> 00:17:09.930 align:middle line:84% about the tax and level of public goods, and then you move. 00:17:09.930 --> 00:17:11.720 align:middle line:84% So there's some bound over which they can 00:17:11.720 --> 00:17:12.970 align:middle line:90% deviate from your preferences. 00:17:12.970 --> 00:17:15.829 align:middle line:84% And as long as it doesn't get too far from your preferences, 00:17:15.829 --> 00:17:16.650 align:middle line:90% you'll stay. 00:17:16.650 --> 00:17:17.349 align:middle line:90% What else? 00:17:17.349 --> 00:17:23.589 align:middle line:90% 00:17:23.589 --> 00:17:24.089 align:middle line:90% What else? 00:17:24.089 --> 00:17:25.238 align:middle line:90% Yeah? 00:17:25.238 --> 00:17:26.530 align:middle line:90% AUDIENCE: Information friction. 00:17:26.530 --> 00:17:28.910 align:middle line:84% Like for example, you might not know [INAUDIBLE]. 00:17:28.910 --> 00:17:30.350 align:middle line:90% JON GRUBER: Exactly. 00:17:30.350 --> 00:17:32.173 align:middle line:84% Information frictions is exactly right. 00:17:32.173 --> 00:17:34.090 align:middle line:84% You don't know what the public librarians pay. 00:17:34.090 --> 00:17:34.987 align:middle line:90% You know your taxes. 00:17:34.987 --> 00:17:37.070 align:middle line:84% But then they stand up and tell me, say, oh, those 00:17:37.070 --> 00:17:38.630 align:middle line:84% taxes supporting vital public goods. 00:17:38.630 --> 00:17:39.850 align:middle line:90% Bullshit, bullshit, bullshit. 00:17:39.850 --> 00:17:40.953 align:middle line:90% And you don't know. 00:17:40.953 --> 00:17:42.370 align:middle line:84% So it's hard to actually know what 00:17:42.370 --> 00:17:44.230 align:middle line:84% your town's doing if they're wasting the money or not. 00:17:44.230 --> 00:17:44.970 align:middle line:90% Yeah? 00:17:44.970 --> 00:17:48.050 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:17:48.050 --> 00:17:49.150 align:middle line:90% JON GRUBER: Yeah. 00:17:49.150 --> 00:17:52.530 align:middle line:84% So basically, here's the thing, there's 00:17:52.530 --> 00:17:56.290 align:middle line:84% a fundamental tension in this model, which 00:17:56.290 --> 00:17:59.740 align:middle line:84% is, on the one hand, why couldn't you 00:17:59.740 --> 00:18:01.990 align:middle line:84% just have-- why do you need a limited number of towns? 00:18:01.990 --> 00:18:04.690 align:middle line:84% Why couldn't there just be each person could be a town? 00:18:04.690 --> 00:18:05.710 align:middle line:90% What's wrong with that? 00:18:05.710 --> 00:18:07.210 align:middle line:84% AUDIENCE: There's some things that are [? too ?] 00:18:07.210 --> 00:18:08.088 align:middle line:90% [? expensive. ?] 00:18:08.088 --> 00:18:09.630 align:middle line:84% JON GRUBER: And what do we call that? 00:18:09.630 --> 00:18:10.650 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:18:10.650 --> 00:18:12.130 align:middle line:84% JON GRUBER: An economies of scale. 00:18:12.130 --> 00:18:14.910 align:middle line:84% Because there's economies of scale in public goods provision. 00:18:14.910 --> 00:18:17.970 align:middle line:84% It makes no sense for you to have your own school. 00:18:17.970 --> 00:18:20.730 align:middle line:84% As a result, it only makes sense to have 00:18:20.730 --> 00:18:23.570 align:middle line:84% schools over some aggregated set of people. 00:18:23.570 --> 00:18:27.970 align:middle line:84% So on the one hand, you have economies of scale. 00:18:27.970 --> 00:18:30.410 align:middle line:84% But on the other hand, if you have economies 00:18:30.410 --> 00:18:34.900 align:middle line:90% of scale versus heterogeneity. 00:18:34.900 --> 00:18:40.220 align:middle line:84% So what Tiebout is assuming is that there 00:18:40.220 --> 00:18:43.420 align:middle line:84% are N identical people that can fit in your town, 00:18:43.420 --> 00:18:45.000 align:middle line:84% that there's enough identical people, 00:18:45.000 --> 00:18:46.560 align:middle line:84% you can afford to build a school, 00:18:46.560 --> 00:18:48.460 align:middle line:84% but they still have identical preferences. 00:18:48.460 --> 00:18:50.473 align:middle line:84% But that's never going to be true. 00:18:50.473 --> 00:18:52.140 align:middle line:84% I mean, maybe you can find-- but I mean, 00:18:52.140 --> 00:18:55.300 align:middle line:84% actually, I don't think people could be homogeneous. 00:18:55.300 --> 00:18:58.403 align:middle line:84% You could form the skydiving town. 00:18:58.403 --> 00:18:59.820 align:middle line:84% And maybe there's enough skydivers 00:18:59.820 --> 00:19:01.340 align:middle line:84% that that's the public good they want to provide 00:19:01.340 --> 00:19:02.560 align:middle line:90% is parachutes for everyone. 00:19:02.560 --> 00:19:03.825 align:middle line:90% And maybe that would work. 00:19:03.825 --> 00:19:05.700 align:middle line:84% But the point is you need a big enough number 00:19:05.700 --> 00:19:08.800 align:middle line:84% of people versus people of similar enough tastes. 00:19:08.800 --> 00:19:09.960 align:middle line:90% And that's attention. 00:19:09.960 --> 00:19:11.325 align:middle line:90% Yeah? 00:19:11.325 --> 00:19:12.283 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:19:12.283 --> 00:19:13.200 align:middle line:90% JON GRUBER: That's OK. 00:19:13.200 --> 00:19:13.860 align:middle line:90% No worries. 00:19:13.860 --> 00:19:15.020 align:middle line:90% Yeah, Steven? 00:19:15.020 --> 00:19:16.580 align:middle line:84% AUDIENCE: Like a public good product 00:19:16.580 --> 00:19:19.180 align:middle line:84% that is being sold by each municipality is not 00:19:19.180 --> 00:19:22.720 align:middle line:84% like homogeneous enough across municipalities [INAUDIBLE]. 00:19:22.720 --> 00:19:25.540 align:middle line:84% JON GRUBER: Yeah, so there's in some sense imperfect 00:19:25.540 --> 00:19:27.620 align:middle line:90% competition. 00:19:27.620 --> 00:19:31.680 align:middle line:84% Some of that comes from costly moving, but some of that 00:19:31.680 --> 00:19:33.320 align:middle line:84% also comes from the fact that there's 00:19:33.320 --> 00:19:34.888 align:middle line:90% a bundle of public goods. 00:19:34.888 --> 00:19:36.680 align:middle line:84% So you need not only to have heterogeneity, 00:19:36.680 --> 00:19:39.880 align:middle line:84% heterogeneity among every single public good. 00:19:39.880 --> 00:19:41.260 align:middle line:84% So that's going to be impossible. 00:19:41.260 --> 00:19:44.320 align:middle line:84% So that's one class of problems with this model. 00:19:44.320 --> 00:19:46.540 align:middle line:84% There's another class of problems, 00:19:46.540 --> 00:19:49.920 align:middle line:84% which is separate but very important, 00:19:49.920 --> 00:19:54.120 align:middle line:84% which is we don't pay for public goods in this way, 00:19:54.120 --> 00:19:55.220 align:middle line:90% these lump sum charges. 00:19:55.220 --> 00:19:57.920 align:middle line:84% That's not how public goods are paid for. 00:19:57.920 --> 00:20:02.840 align:middle line:84% Lump sum tax, it turns out, are incredibly unpopular. 00:20:02.840 --> 00:20:05.340 align:middle line:84% Actually, I'll talk about this when we talk about taxation. 00:20:05.340 --> 00:20:06.800 align:middle line:84% But basically, the only modern example 00:20:06.800 --> 00:20:08.600 align:middle line:84% of a government trying to impose lump sum taxes 00:20:08.600 --> 00:20:10.880 align:middle line:84% was Margaret Thatcher, who was the most popular leader 00:20:10.880 --> 00:20:13.480 align:middle line:84% in modern history, proposed a lump sum tax in England 00:20:13.480 --> 00:20:16.585 align:middle line:84% and lost office over it because people view them as unfair, 00:20:16.585 --> 00:20:18.960 align:middle line:84% fundamentally viewed unfair that Bill Gates and you would 00:20:18.960 --> 00:20:22.440 align:middle line:90% pay the same tax. 00:20:22.440 --> 00:20:25.920 align:middle line:84% As a result, nobody fancies anything through lump sum taxes. 00:20:25.920 --> 00:20:30.740 align:middle line:84% Typically, how our local public goods financed through property 00:20:30.740 --> 00:20:33.340 align:middle line:90% taxes. 00:20:33.340 --> 00:20:36.340 align:middle line:84% That is the way that local public goods are financed are 00:20:36.340 --> 00:20:39.020 align:middle line:84% through a tax on property, primarily housing, 00:20:39.020 --> 00:20:41.360 align:middle line:84% 90-plus percent of property tax revenues. 00:20:41.360 --> 00:20:42.958 align:middle line:84% There's some other property taxes, 00:20:42.958 --> 00:20:44.500 align:middle line:84% but basically, it's taxes on housing. 00:20:44.500 --> 00:20:47.580 align:middle line:84% So you can think of public goods at the local level 00:20:47.580 --> 00:20:51.980 align:middle line:84% as being paid for by a tax on housing. 00:20:51.980 --> 00:20:55.340 align:middle line:84% Why does that screw up this model? 00:20:55.340 --> 00:20:57.740 align:middle line:90% Well, here's a question for you. 00:20:57.740 --> 00:21:01.190 align:middle line:90% Let's say you're a rich person. 00:21:01.190 --> 00:21:03.930 align:middle line:84% And you know that public goods are financed-- 00:21:03.930 --> 00:21:05.180 align:middle line:90% rich people have fancy houses. 00:21:05.180 --> 00:21:08.880 align:middle line:84% You know that public goods are financed by property tax. 00:21:08.880 --> 00:21:11.020 align:middle line:84% Who else do you want to live with? 00:21:11.020 --> 00:21:12.360 align:middle line:90% Alec? 00:21:12.360 --> 00:21:13.380 align:middle line:84% AUDIENCE: Probably a bunch of other rich people. 00:21:13.380 --> 00:21:14.580 align:middle line:84% JON GRUBER: Probably a bunch of other people. 00:21:14.580 --> 00:21:16.160 align:middle line:84% Now, let's say you're a poor person. 00:21:16.160 --> 00:21:17.360 align:middle line:84% Alec, who do you want to live with? 00:21:17.360 --> 00:21:18.220 align:middle line:90% AUDIENCE: [INAUDIBLE] people. 00:21:18.220 --> 00:21:20.180 align:middle line:84% JON GRUBER: A bunch of other rich people. 00:21:20.180 --> 00:21:22.800 align:middle line:84% The problem is, in a world with property taxes, 00:21:22.800 --> 00:21:25.380 align:middle line:84% you have a bit of a free rider problem 00:21:25.380 --> 00:21:27.043 align:middle line:84% because basically, the rich pay more, 00:21:27.043 --> 00:21:28.460 align:middle line:84% but everyone gets the public good. 00:21:28.460 --> 00:21:30.980 align:middle line:84% And if they're not all homogeneous, that screws it up. 00:21:30.980 --> 00:21:33.760 align:middle line:84% So heterogeneity in income with a property tax 00:21:33.760 --> 00:21:35.800 align:middle line:90% means this doesn't work. 00:21:35.800 --> 00:21:37.780 align:middle line:84% But the rich have found a solution to this. 00:21:37.780 --> 00:21:40.800 align:middle line:90% And what's that called? 00:21:40.800 --> 00:21:43.000 align:middle line:84% How do the rich solve the free rider problem 00:21:43.000 --> 00:21:45.962 align:middle line:90% inherent in this system? 00:21:45.962 --> 00:21:47.420 align:middle line:84% They claim it's for something else. 00:21:47.420 --> 00:21:48.170 align:middle line:90% What is it really? 00:21:48.170 --> 00:21:49.320 align:middle line:90% Yeah? 00:21:49.320 --> 00:21:51.282 align:middle line:84% AUDIENCE: I mean, they provide private goods. 00:21:51.282 --> 00:21:53.240 align:middle line:84% JON GRUBER: Well, private goods, but what else? 00:21:53.240 --> 00:21:53.960 align:middle line:90% Yeah? 00:21:53.960 --> 00:21:56.640 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:21:56.640 --> 00:21:58.000 align:middle line:90% JON GRUBER: [INAUDIBLE] what? 00:21:58.000 --> 00:22:06.280 align:middle line:84% AUDIENCE: [INAUDIBLE] you can exclude people [INAUDIBLE]. 00:22:06.280 --> 00:22:10.700 align:middle line:84% JON GRUBER: That's typically illegal to do that, not always. 00:22:10.700 --> 00:22:12.840 align:middle line:90% What else do towns do-- 00:22:12.840 --> 00:22:15.360 align:middle line:84% what do towns do to try to fight the fact 00:22:15.360 --> 00:22:19.620 align:middle line:84% that both the rich and the poor want to live with the rich? 00:22:19.620 --> 00:22:20.120 align:middle line:90% Yeah? 00:22:20.120 --> 00:22:21.060 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:22:21.060 --> 00:22:22.840 align:middle line:90% JON GRUBER: Zoning laws. 00:22:22.840 --> 00:22:26.900 align:middle line:84% So Western Massachusetts will tell you 00:22:26.900 --> 00:22:29.612 align:middle line:84% that it restricts everyone to have at least two acres 00:22:29.612 --> 00:22:31.320 align:middle line:84% because it wants to preserve green space, 00:22:31.320 --> 00:22:32.320 align:middle line:90% but that's bullshit. 00:22:32.320 --> 00:22:33.755 align:middle line:84% They want more than two two acres 00:22:33.755 --> 00:22:35.380 align:middle line:84% because that keeps the poor people out. 00:22:35.380 --> 00:22:37.700 align:middle line:84% Poor people can't afford two acres. 00:22:37.700 --> 00:22:42.460 align:middle line:84% So zoning laws are how towns have reacted to the incentives 00:22:42.460 --> 00:22:44.620 align:middle line:90% to free ride on property taxes. 00:22:44.620 --> 00:22:49.867 align:middle line:84% When you have a property tax where the rich pay more, 00:22:49.867 --> 00:22:51.700 align:middle line:84% if you don't keep the poor out, then they're 00:22:51.700 --> 00:22:54.060 align:middle line:84% going to free ride on the taxes the rich pay. 00:22:54.060 --> 00:22:56.780 align:middle line:84% So that's why towns have zoning laws. 00:22:56.780 --> 00:22:59.200 align:middle line:84% These zoning laws are incredibly pernicious. 00:22:59.200 --> 00:23:01.500 align:middle line:84% They're, first of all, incredibly prevalent, but also 00:23:01.500 --> 00:23:03.292 align:middle line:84% incredibly pernicious, because what they do 00:23:03.292 --> 00:23:07.420 align:middle line:84% is they massively restrict the amount of housing available. 00:23:07.420 --> 00:23:09.165 align:middle line:84% Because basically, the way you do zoning, 00:23:09.165 --> 00:23:10.540 align:middle line:84% you can't say a house in our town 00:23:10.540 --> 00:23:11.880 align:middle line:90% has to cost at least $1 million. 00:23:11.880 --> 00:23:12.770 align:middle line:90% That's illegal. 00:23:12.770 --> 00:23:15.020 align:middle line:84% But you can say it has to at least have amenities such 00:23:15.020 --> 00:23:17.940 align:middle line:84% that no one would build such a house for less than $1 million. 00:23:17.940 --> 00:23:19.565 align:middle line:84% And those amenities mean that you don't 00:23:19.565 --> 00:23:20.920 align:middle line:90% get many houses in your town. 00:23:20.920 --> 00:23:21.420 align:middle line:90% Yeah? 00:23:21.420 --> 00:23:23.740 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:23:23.740 --> 00:23:30.425 align:middle line:90% 00:23:30.425 --> 00:23:33.050 align:middle line:84% JON GRUBER: Yeah, well, we have two things like that in the US. 00:23:33.050 --> 00:23:35.657 align:middle line:84% We have within buildings, what we call co-op buildings, which 00:23:35.657 --> 00:23:37.990 align:middle line:84% are essentially cooperatives where the other owners have 00:23:37.990 --> 00:23:38.970 align:middle line:90% to prove you. 00:23:38.970 --> 00:23:41.410 align:middle line:84% And if you want to make a renovation in your apartment, 00:23:41.410 --> 00:23:43.310 align:middle line:90% the whole co-op has to approve. 00:23:43.310 --> 00:23:46.930 align:middle line:84% We also have some communities-- there's a community behind me, 00:23:46.930 --> 00:23:51.710 align:middle line:84% which was set up as a community with what's called a covenant. 00:23:51.710 --> 00:23:54.635 align:middle line:84% And basically, what that means is not anyone can buy in, 00:23:54.635 --> 00:23:56.510 align:middle line:84% but you can't, for example, change your house 00:23:56.510 --> 00:23:59.350 align:middle line:84% without approval of everyone in the neighborhood. 00:23:59.350 --> 00:24:01.430 align:middle line:84% But you can't stop someone from buying in, 00:24:01.430 --> 00:24:03.970 align:middle line:84% certainly not for any reason like they're poor. 00:24:03.970 --> 00:24:05.830 align:middle line:90% You get sued for that. 00:24:05.830 --> 00:24:09.630 align:middle line:90% But zoning, totally fine. 00:24:09.630 --> 00:24:11.990 align:middle line:84% The effects of zoning are enormous. 00:24:11.990 --> 00:24:17.310 align:middle line:84% The best estimates-- there's an excellent urban economist 00:24:17.310 --> 00:24:19.070 align:middle line:90% named Ed Glaeser at Harvard. 00:24:19.070 --> 00:24:25.573 align:middle line:84% They said that land in zoned areas costs about 10 times more 00:24:25.573 --> 00:24:27.990 align:middle line:84% than the land in the unzoned areas, just roughly speaking, 00:24:27.990 --> 00:24:30.170 align:middle line:90% to try to compare them. 00:24:30.170 --> 00:24:34.390 align:middle line:84% Zoning is a huge, huge effect on the cost of housing, 00:24:34.390 --> 00:24:36.410 align:middle line:90% availability of housing. 00:24:36.410 --> 00:24:41.730 align:middle line:84% But it's a natural reaction to the Tiebout mechanism. 00:24:41.730 --> 00:24:43.970 align:middle line:84% So in other words, competition-- think of it 00:24:43.970 --> 00:24:45.492 align:middle line:90% like we did in 14.01. 00:24:45.492 --> 00:24:47.450 align:middle line:84% Competition only delivers the efficient outcome 00:24:47.450 --> 00:24:49.870 align:middle line:84% if there's no barriers. some mentioned barriers to entry. 00:24:49.870 --> 00:24:51.650 align:middle line:84% Well, here's a barrier that makes competition not 00:24:51.650 --> 00:24:52.900 align:middle line:90% deliver the efficient outcome. 00:24:52.900 --> 00:24:53.400 align:middle line:90% Yeah? 00:24:53.400 --> 00:24:54.890 align:middle line:84% AUDIENCE: Would the zoning thing be 00:24:54.890 --> 00:24:58.730 align:middle line:90% different than the [INAUDIBLE] 00:24:58.730 --> 00:25:00.890 align:middle line:84% JON GRUBER: That's about getting your voters. 00:25:00.890 --> 00:25:02.770 align:middle line:84% This is about literally who can live there. 00:25:02.770 --> 00:25:04.145 align:middle line:84% AUDIENCE: Districting [INAUDIBLE] 00:25:04.145 --> 00:25:06.033 align:middle line:84% like school districts [INAUDIBLE] 00:25:06.033 --> 00:25:06.950 align:middle line:90% JON GRUBER: Oh, I see. 00:25:06.950 --> 00:25:09.170 align:middle line:84% Yes, sometimes they have school districts that 00:25:09.170 --> 00:25:10.748 align:middle line:90% don't follow town boundaries. 00:25:10.748 --> 00:25:11.790 align:middle line:90% And that's another thing. 00:25:11.790 --> 00:25:14.410 align:middle line:84% But that's more about economies of scale. 00:25:14.410 --> 00:25:16.845 align:middle line:90% Now there's a third problem. 00:25:16.845 --> 00:25:18.470 align:middle line:84% So there's three problems with Tiebout. 00:25:18.470 --> 00:25:21.090 align:middle line:90% One is market imperfections. 00:25:21.090 --> 00:25:23.750 align:middle line:90% Two is the financing. 00:25:23.750 --> 00:25:28.230 align:middle line:84% And then this third problem with Tiebout, which is it 00:25:28.230 --> 00:25:32.910 align:middle line:90% assumes no spillovers. 00:25:32.910 --> 00:25:33.770 align:middle line:90% But that's not true. 00:25:33.770 --> 00:25:37.690 align:middle line:84% Anytime a town builds public goods, it affects other towns. 00:25:37.690 --> 00:25:40.390 align:middle line:84% If I paved my road, then people who drive through 00:25:40.390 --> 00:25:42.250 align:middle line:90% on their way home benefit. 00:25:42.250 --> 00:25:45.165 align:middle line:84% If I have better police, then there's less crime. 00:25:45.165 --> 00:25:47.790 align:middle line:84% Maybe that hurts the other town because it pushes the criminals 00:25:47.790 --> 00:25:48.572 align:middle line:90% to the other town. 00:25:48.572 --> 00:25:50.030 align:middle line:84% If I have better education, there's 00:25:50.030 --> 00:25:51.447 align:middle line:84% better educated workers that maybe 00:25:51.447 --> 00:25:52.990 align:middle line:90% want to come work in my town. 00:25:52.990 --> 00:25:56.230 align:middle line:84% Once there's spillovers, that leads to the same market failure 00:25:56.230 --> 00:25:57.670 align:middle line:84% we get with positive externalities 00:25:57.670 --> 00:26:00.310 align:middle line:84% or negative externalities in our standard market, 00:26:00.310 --> 00:26:03.310 align:middle line:84% which is competition doesn't work if there's externalities. 00:26:03.310 --> 00:26:06.180 align:middle line:84% Well, with spillovers in public goods, there's externalities. 00:26:06.180 --> 00:26:07.930 align:middle line:84% Therefore, we won't get the optimal level. 00:26:07.930 --> 00:26:11.270 align:middle line:84% But even if we don't worry with these other conditions, 00:26:11.270 --> 00:26:13.630 align:middle line:84% Tiebout model will not deliver optimality 00:26:13.630 --> 00:26:18.890 align:middle line:84% because we will not get the optimal level of public goods 00:26:18.890 --> 00:26:22.330 align:middle line:84% because the spillovers are externalities. 00:26:22.330 --> 00:26:24.670 align:middle line:84% OK, questions about the Tiebout model? 00:26:24.670 --> 00:26:27.450 align:middle line:90% 00:26:27.450 --> 00:26:32.010 align:middle line:84% Now, much like Coase or the median voter theorem, 00:26:32.010 --> 00:26:34.670 align:middle line:84% the Tiebout model is not true in extreme, 00:26:34.670 --> 00:26:37.330 align:middle line:84% but has a lot of power to explain phenomena 00:26:37.330 --> 00:26:38.930 align:middle line:90% we see in the real world. 00:26:38.930 --> 00:26:41.690 align:middle line:84% So what are some evidence that the Tiebout model actually 00:26:41.690 --> 00:26:43.130 align:middle line:90% works in the real world. 00:26:43.130 --> 00:26:48.530 align:middle line:84% Well, for example, let's imagine-- 00:26:48.530 --> 00:26:52.410 align:middle line:84% let's imagine two cities, one of which 00:26:52.410 --> 00:26:59.490 align:middle line:84% has three big suburbs, one of which has 20 little suburbs. 00:26:59.490 --> 00:27:02.090 align:middle line:84% What does the Tiebout model predict 00:27:02.090 --> 00:27:04.710 align:middle line:84% should be different about the suburbs in those two cities? 00:27:04.710 --> 00:27:08.000 align:middle line:90% 00:27:08.000 --> 00:27:10.430 align:middle line:90% What things-- not one thing. 00:27:10.430 --> 00:27:11.778 align:middle line:90% What things should be different? 00:27:11.778 --> 00:27:13.570 align:middle line:84% Going to try some other folks involved now. 00:27:13.570 --> 00:27:13.910 align:middle line:90% Let's see. 00:27:13.910 --> 00:27:14.190 align:middle line:90% Come on. 00:27:14.190 --> 00:27:15.690 align:middle line:84% Let's get some other folks involved. 00:27:15.690 --> 00:27:17.190 align:middle line:90% Let's get more participation up. 00:27:17.190 --> 00:27:18.870 align:middle line:84% What other things should be different? 00:27:18.870 --> 00:27:22.250 align:middle line:84% They want a place with a few or even one to make it extreme, 00:27:22.250 --> 00:27:24.790 align:middle line:84% one big suburb versus 20 little suburbs. 00:27:24.790 --> 00:27:27.070 align:middle line:84% What should be different in the nature of the suburbs 00:27:27.070 --> 00:27:29.842 align:middle line:90% in those two places? 00:27:29.842 --> 00:27:32.050 align:middle line:84% Actually, one doesn't work, got to be at least three. 00:27:32.050 --> 00:27:32.810 align:middle line:90% Yeah? 00:27:32.810 --> 00:27:34.602 align:middle line:84% AUDIENCE: In larger cases, there would more 00:27:34.602 --> 00:27:36.870 align:middle line:84% specialization in public goods or more [INAUDIBLE] 00:27:36.870 --> 00:27:39.770 align:middle line:84% JON GRUBER: With more towns, there'd be more specialization. 00:27:39.770 --> 00:27:40.338 align:middle line:90% Absolutely. 00:27:40.338 --> 00:27:42.130 align:middle line:84% You predict there'd be more specialization, 00:27:42.130 --> 00:27:44.330 align:middle line:84% there'd be more diversity across towns. 00:27:44.330 --> 00:27:44.830 align:middle line:90% Yeah? 00:27:44.830 --> 00:27:47.470 align:middle line:84% AUDIENCE: People in towns would be more similar to one another. 00:27:47.470 --> 00:27:48.303 align:middle line:90% JON GRUBER: Exactly. 00:27:48.303 --> 00:27:50.410 align:middle line:84% You'd expect more-- those are the two predictions. 00:27:50.410 --> 00:27:52.790 align:middle line:84% You'd expect more diversity in public goods provision 00:27:52.790 --> 00:27:55.150 align:middle line:84% as the number of towns goes up and more 00:27:55.150 --> 00:27:57.570 align:middle line:84% homogeneity within a town the number of towns go up. 00:27:57.570 --> 00:27:59.630 align:middle line:90% And both are true. 00:27:59.630 --> 00:28:00.810 align:middle line:90% It is true. 00:28:00.810 --> 00:28:04.030 align:middle line:84% But if you think about 15 versus 20, as the number gets bigger, 00:28:04.030 --> 00:28:06.355 align:middle line:84% you get more specialization and more homogeneity. 00:28:06.355 --> 00:28:08.230 align:middle line:84% Never all the way to what Tiebout would want. 00:28:08.230 --> 00:28:11.630 align:middle line:84% But you go in that direction as predicted by that model. 00:28:11.630 --> 00:28:15.350 align:middle line:84% You also get a higher level of satisfaction with public goods 00:28:15.350 --> 00:28:17.123 align:middle line:84% as the number of towns get bigger 00:28:17.123 --> 00:28:19.040 align:middle line:84% because people are moving towards that Tiebout 00:28:19.040 --> 00:28:19.960 align:middle line:90% equilibrium. 00:28:19.960 --> 00:28:21.960 align:middle line:84% So the more choice they have, the happier people 00:28:21.960 --> 00:28:23.960 align:middle line:84% are because they can move to a place that really 00:28:23.960 --> 00:28:26.000 align:middle line:90% matches their preferences. 00:28:26.000 --> 00:28:28.840 align:middle line:90% So that's one prediction. 00:28:28.840 --> 00:28:31.600 align:middle line:84% The second prediction is really important and interesting 00:28:31.600 --> 00:28:36.307 align:middle line:84% because another problem with Tiebout model 00:28:36.307 --> 00:28:38.640 align:middle line:84% is that-- and someone mentioned this in the first thing. 00:28:38.640 --> 00:28:40.680 align:middle line:90% Moving is a big deal. 00:28:40.680 --> 00:28:42.060 align:middle line:90% It's really costly to move. 00:28:42.060 --> 00:28:43.480 align:middle line:84% I mean, you guys move your dorm rooms every year 00:28:43.480 --> 00:28:44.480 align:middle line:90% and you don't care. 00:28:44.480 --> 00:28:47.700 align:middle line:84% But believe me, like selling a house, buying a new house, 00:28:47.700 --> 00:28:49.160 align:middle line:90% it's a nightmare. 00:28:49.160 --> 00:28:51.025 align:middle line:90% It's really costly. 00:28:51.025 --> 00:28:52.400 align:middle line:84% And so one problem you might have 00:28:52.400 --> 00:28:54.860 align:middle line:84% with the gee, Tiebout model, gee, 00:28:54.860 --> 00:28:56.140 align:middle line:90% all these people have to move. 00:28:56.140 --> 00:28:57.625 align:middle line:90% That's a very weak mechanism. 00:28:57.625 --> 00:28:58.500 align:middle line:90% But here's the thing. 00:28:58.500 --> 00:29:00.680 align:middle line:90% They don't have to move. 00:29:00.680 --> 00:29:03.000 align:middle line:84% You don't need that many people to move. 00:29:03.000 --> 00:29:05.880 align:middle line:84% In fact, in theory, you don't need anyone to move 00:29:05.880 --> 00:29:08.260 align:middle line:84% or certainly a very small number of people to move. 00:29:08.260 --> 00:29:10.120 align:middle line:90% And why is that? 00:29:10.120 --> 00:29:13.460 align:middle line:84% Why would the Tiebout force operate in theory 00:29:13.460 --> 00:29:15.600 align:middle line:84% even if no one moves, but as long as you 00:29:15.600 --> 00:29:18.015 align:middle line:90% don't need many people to move? 00:29:18.015 --> 00:29:20.140 align:middle line:84% What else is going to change is to make the Tiebout 00:29:20.140 --> 00:29:21.660 align:middle line:90% force operate? 00:29:21.660 --> 00:29:23.980 align:middle line:90% You're the mayor of a town. 00:29:23.980 --> 00:29:26.605 align:middle line:90% You've screwed up. 00:29:26.605 --> 00:29:27.980 align:middle line:84% How are you going to get a signal 00:29:27.980 --> 00:29:29.260 align:middle line:90% without everyone leaving town? 00:29:29.260 --> 00:29:31.260 align:middle line:84% How are you going to signal that you screwed up? 00:29:31.260 --> 00:29:33.320 align:middle line:84% What's going to tell you you screwed up? 00:29:33.320 --> 00:29:33.820 align:middle line:90% Yeah? 00:29:33.820 --> 00:29:34.420 align:middle line:90% What? 00:29:34.420 --> 00:29:34.840 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:29:34.840 --> 00:29:35.548 align:middle line:90% JON GRUBER: What? 00:29:35.548 --> 00:29:36.760 align:middle line:90% AUDIENCE: Get voted out. 00:29:36.760 --> 00:29:38.760 align:middle line:84% JON GRUBER: But there's no vote for three years. 00:29:38.760 --> 00:29:39.260 align:middle line:90% Yeah? 00:29:39.260 --> 00:29:44.080 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:29:44.080 --> 00:29:46.380 align:middle line:84% JON GRUBER: But let's imagine you throw them in jail. 00:29:46.380 --> 00:29:47.900 align:middle line:90% OK, what else? 00:29:47.900 --> 00:29:49.940 align:middle line:84% What else can you observe in economic indicators 00:29:49.940 --> 00:29:50.960 align:middle line:90% about that town? 00:29:50.960 --> 00:29:51.460 align:middle line:90% Yeah? 00:29:51.460 --> 00:29:53.495 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:29:53.495 --> 00:29:55.120 align:middle line:84% JON GRUBER: Well, that's what she said. 00:29:55.120 --> 00:29:56.280 align:middle line:90% But then you throw them in jail. 00:29:56.280 --> 00:29:56.920 align:middle line:90% What else? 00:29:56.920 --> 00:29:59.740 align:middle line:90% 00:29:59.740 --> 00:30:02.300 align:middle line:84% Think about how public goods are financed. 00:30:02.300 --> 00:30:03.960 align:middle line:90% Think about markets and towns. 00:30:03.960 --> 00:30:05.180 align:middle line:90% What's going to change? 00:30:05.180 --> 00:30:05.698 align:middle line:90% Yeah? 00:30:05.698 --> 00:30:07.740 align:middle line:84% AUDIENCE: You'll see more private good provision. 00:30:07.740 --> 00:30:09.740 align:middle line:84% JON GRUBER: Private good provision is one thing. 00:30:09.740 --> 00:30:11.340 align:middle line:90% What else? 00:30:11.340 --> 00:30:14.680 align:middle line:84% What about the fundamental nature of prices in the economy 00:30:14.680 --> 00:30:16.390 align:middle line:90% will change? 00:30:16.390 --> 00:30:17.330 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:30:17.330 --> 00:30:18.330 align:middle line:90% JON GRUBER: What's that? 00:30:18.330 --> 00:30:20.280 align:middle line:84% AUDIENCE: The cost of the [INAUDIBLE] 00:30:20.280 --> 00:30:23.160 align:middle line:84% JON GRUBER: House prices will drop because people don't 00:30:23.160 --> 00:30:24.520 align:middle line:90% want to live there anymore. 00:30:24.520 --> 00:30:27.778 align:middle line:84% And asset prices can react much quicker than people can move. 00:30:27.778 --> 00:30:29.820 align:middle line:84% So in theory, you don't even need anyone to move. 00:30:29.820 --> 00:30:33.120 align:middle line:84% If it gets in the news that this town sucks, all of a sudden, 00:30:33.120 --> 00:30:35.540 align:middle line:84% no one wants to live there, and house prices drop. 00:30:35.540 --> 00:30:37.440 align:middle line:84% You don't actually even need mobility. 00:30:37.440 --> 00:30:39.560 align:middle line:90% That's what's great about this. 00:30:39.560 --> 00:30:42.200 align:middle line:84% This is what we call capitalization. 00:30:42.200 --> 00:30:44.600 align:middle line:84% There's something called capitalization, 00:30:44.600 --> 00:30:47.880 align:middle line:84% which is the notion that a house price embeds 00:30:47.880 --> 00:30:50.320 align:middle line:84% a stream of services that come with that house. 00:30:50.320 --> 00:30:52.720 align:middle line:84% Why is one house expensive, one house not expensive? 00:30:52.720 --> 00:30:54.720 align:middle line:84% Well, it embeds how many bedrooms there are 00:30:54.720 --> 00:30:56.128 align:middle line:90% and how nice it is in the yard. 00:30:56.128 --> 00:30:57.920 align:middle line:84% But it also embeds how good the schools are 00:30:57.920 --> 00:30:59.960 align:middle line:84% and how high the property taxes are. 00:30:59.960 --> 00:31:02.520 align:middle line:84% Those are all embedded in one stream of things that 00:31:02.520 --> 00:31:04.480 align:middle line:90% determine the price of a house. 00:31:04.480 --> 00:31:07.360 align:middle line:84% So for a given level of property taxes, 00:31:07.360 --> 00:31:09.200 align:middle line:84% if I suddenly start keeping all the money 00:31:09.200 --> 00:31:11.350 align:middle line:84% for my public librarian sports car 00:31:11.350 --> 00:31:13.600 align:middle line:84% and don't spend it on things that are good for people, 00:31:13.600 --> 00:31:15.517 align:middle line:84% they'll say, don't live here, this town sucks. 00:31:15.517 --> 00:31:17.660 align:middle line:90% And house prices will drop. 00:31:17.660 --> 00:31:22.300 align:middle line:84% So you don't need everyone to move to have the market 00:31:22.300 --> 00:31:26.540 align:middle line:90% show that you've screwed up. 00:31:26.540 --> 00:31:29.020 align:middle line:84% So what this says-- what's really cool about this 00:31:29.020 --> 00:31:31.660 align:middle line:84% is it says that even in a world where people don't move much, 00:31:31.660 --> 00:31:35.980 align:middle line:84% we can use house prices as an indicator for how good 00:31:35.980 --> 00:31:38.460 align:middle line:84% a job a government is doing on meeting 00:31:38.460 --> 00:31:41.620 align:middle line:90% the tastes of its citizens. 00:31:41.620 --> 00:31:44.940 align:middle line:84% So in other words, if you cut taxes, 00:31:44.940 --> 00:31:47.300 align:middle line:90% if you cut property taxes-- 00:31:47.300 --> 00:31:51.420 align:middle line:84% let's say you have a town, and you cut property taxes. 00:31:51.420 --> 00:31:54.300 align:middle line:84% Should house prices go down or up? 00:31:54.300 --> 00:31:56.100 align:middle line:90% And for what reasons? 00:31:56.100 --> 00:31:58.520 align:middle line:84% If you cut property taxes in your town, 00:31:58.520 --> 00:32:02.140 align:middle line:84% should the price of the house go down or even up? 00:32:02.140 --> 00:32:02.800 align:middle line:90% And why? 00:32:02.800 --> 00:32:04.900 align:middle line:90% Yeah? 00:32:04.900 --> 00:32:06.900 align:middle line:84% AUDIENCE: I think they'll go up because that 00:32:06.900 --> 00:32:08.913 align:middle line:90% makes it a more desirable town. 00:32:08.913 --> 00:32:10.580 align:middle line:84% JON GRUBER: You pay less taxes, but does 00:32:10.580 --> 00:32:11.830 align:middle line:90% it make a more desirable town? 00:32:11.830 --> 00:32:14.940 align:middle line:84% AUDIENCE: [INAUDIBLE] people want more public goods. 00:32:14.940 --> 00:32:15.440 align:middle line:90% [INAUDIBLE] 00:32:15.440 --> 00:32:16.760 align:middle line:90% JON GRUBER: Yeah. 00:32:16.760 --> 00:32:19.340 align:middle line:84% In fact, if you're at the Tiebout equilibrium, 00:32:19.340 --> 00:32:20.540 align:middle line:90% it'll make prices go down. 00:32:20.540 --> 00:32:20.760 align:middle line:90% Why? 00:32:20.760 --> 00:32:22.218 align:middle line:84% Because at the Tiebout equilibrium, 00:32:22.218 --> 00:32:24.000 align:middle line:90% everybody's perfectly happy. 00:32:24.000 --> 00:32:26.580 align:middle line:84% So if I cut property taxes in Tiebout equilibrium, 00:32:26.580 --> 00:32:28.160 align:middle line:90% prices will go down. 00:32:28.160 --> 00:32:32.840 align:middle line:84% But if I'm overspending on public goods, prices will go up. 00:32:32.840 --> 00:32:35.760 align:middle line:84% If I'm underspending on public goods, price will go down. 00:32:35.760 --> 00:32:39.240 align:middle line:84% So what happens to prices can tell us 00:32:39.240 --> 00:32:42.120 align:middle line:84% about where we are relative to Tiebout equilibrium, 00:32:42.120 --> 00:32:43.760 align:middle line:84% assuming people are informed and stuff. 00:32:43.760 --> 00:32:44.100 align:middle line:90% Yeah? 00:32:44.100 --> 00:32:45.892 align:middle line:84% AUDIENCE: [INAUDIBLE] like a positive loop? 00:32:45.892 --> 00:32:48.788 align:middle line:90% 00:32:48.788 --> 00:32:51.160 align:middle line:90% JON GRUBER: No, not necessarily. 00:32:51.160 --> 00:32:52.420 align:middle line:90% Explain what you mean by that. 00:32:52.420 --> 00:32:55.122 align:middle line:84% AUDIENCE: Like if tax go up, price goes [INAUDIBLE]. 00:32:55.122 --> 00:32:56.080 align:middle line:90% JON GRUBER: No, no, no. 00:32:56.080 --> 00:33:00.120 align:middle line:84% No, because as the tax goes up-- so let's say you were 00:33:00.120 --> 00:33:02.160 align:middle line:90% underproviding public goods. 00:33:02.160 --> 00:33:04.440 align:middle line:84% So your tax goes up, price goes up. 00:33:04.440 --> 00:33:06.268 align:middle line:90% But now public goods go up too. 00:33:06.268 --> 00:33:08.060 align:middle line:84% Now let's say you've hit the optimal point. 00:33:08.060 --> 00:33:09.900 align:middle line:84% Well, tax goes up again, price will go down. 00:33:09.900 --> 00:33:11.970 align:middle line:90% Think of it as climbing a hill. 00:33:11.970 --> 00:33:16.410 align:middle line:84% The hill is the net of taxes versus benefits. 00:33:16.410 --> 00:33:18.350 align:middle line:84% The hill is the net of tax versus benefits. 00:33:18.350 --> 00:33:20.770 align:middle line:84% And Tiebout is saying you want to settle at the top 00:33:20.770 --> 00:33:23.130 align:middle line:84% where the next dollar in tax is worth it 00:33:23.130 --> 00:33:25.790 align:middle line:84% in terms of the next dollar of benefits it pays for. 00:33:25.790 --> 00:33:29.130 align:middle line:84% Indeed, what's the key lesson from Tiebout 00:33:29.130 --> 00:33:32.690 align:middle line:84% is that we want to think about tax benefit linkages. 00:33:32.690 --> 00:33:35.490 align:middle line:90% 00:33:35.490 --> 00:33:37.470 align:middle line:84% It's a totally new way of thinking about taxes. 00:33:37.470 --> 00:33:39.137 align:middle line:84% We're going to think about taxes, things 00:33:39.137 --> 00:33:40.710 align:middle line:90% like are they efficient? 00:33:40.710 --> 00:33:42.110 align:middle line:90% Do they help poor people? 00:33:42.110 --> 00:33:44.110 align:middle line:84% There's a different way to think about it, which 00:33:44.110 --> 00:33:46.530 align:middle line:90% is think of a tax as a price. 00:33:46.530 --> 00:33:50.130 align:middle line:84% In Tiebout's world, a local tax is a price at which 00:33:50.130 --> 00:33:52.627 align:middle line:90% you purchase a public good. 00:33:52.627 --> 00:33:54.710 align:middle line:84% And when you think of it that way, it makes sense. 00:33:54.710 --> 00:33:56.690 align:middle line:84% If the price is too high, you buy less. 00:33:56.690 --> 00:33:58.570 align:middle line:90% Price is too low, you buy more. 00:33:58.570 --> 00:34:02.130 align:middle line:84% So that sets the net of the tax versus the benefit you're 00:34:02.130 --> 00:34:03.948 align:middle line:84% getting on the margin determines the value 00:34:03.948 --> 00:34:05.990 align:middle line:84% of the house and your desirable to live in there. 00:34:05.990 --> 00:34:06.490 align:middle line:90% Yeah? 00:34:06.490 --> 00:34:08.750 align:middle line:84% AUDIENCE: This model [INAUDIBLE] for example. 00:34:08.750 --> 00:34:11.510 align:middle line:84% Are we selling this good primarily 00:34:11.510 --> 00:34:14.690 align:middle line:84% to the people who are already [INAUDIBLE] 00:34:14.690 --> 00:34:16.170 align:middle line:84% JON GRUBER: I'm leaving all the-- 00:34:16.170 --> 00:34:18.290 align:middle line:84% once again, we're leaving chapter 9 behind. 00:34:18.290 --> 00:34:22.931 align:middle line:84% We're assuming a social planner who's vote maximized 00:34:22.931 --> 00:34:23.889 align:middle line:90% or something like that. 00:34:23.889 --> 00:34:25.662 align:middle line:84% And we're just doing an easy case. 00:34:25.662 --> 00:34:26.370 align:middle line:90% But you're right. 00:34:26.370 --> 00:34:28.429 align:middle line:84% There'd be a lot of subtleties about pleasing 00:34:28.429 --> 00:34:31.389 align:middle line:84% maybe certain residents of town who hold the power versus not, 00:34:31.389 --> 00:34:33.530 align:middle line:84% et cetera, as we talked about in chapter 9. 00:34:33.530 --> 00:34:34.030 align:middle line:90% Yeah? 00:34:34.030 --> 00:34:35.630 align:middle line:84% AUDIENCE: [INAUDIBLE] capitalization. 00:34:35.630 --> 00:34:37.670 align:middle line:84% Are there any other salient examples 00:34:37.670 --> 00:34:39.852 align:middle line:90% besides like housing prices? 00:34:39.852 --> 00:34:41.310 align:middle line:84% JON GRUBER: Oh, it's in everything. 00:34:41.310 --> 00:34:41.969 align:middle line:90% I mean. 00:34:41.969 --> 00:34:44.310 align:middle line:90% So yeah, the stock market. 00:34:44.310 --> 00:34:46.270 align:middle line:84% What determines the price of a stock? 00:34:46.270 --> 00:34:49.710 align:middle line:84% Well, in theory, it's the underlying future profits 00:34:49.710 --> 00:34:52.449 align:middle line:84% of that company determine the value of a stock. 00:34:52.449 --> 00:34:54.590 align:middle line:84% Now, in practice, it's kind of how day traders are 00:34:54.590 --> 00:34:55.750 align:middle line:90% feeling that day. 00:34:55.750 --> 00:34:59.150 align:middle line:84% But in theory-- so, yes, capitalization is fundamental. 00:34:59.150 --> 00:35:03.070 align:middle line:84% All assets are capitalized something. 00:35:03.070 --> 00:35:06.150 align:middle line:84% Every asset you hold is a capitalized value 00:35:06.150 --> 00:35:10.970 align:middle line:84% of its future utilization or future utility or future value. 00:35:10.970 --> 00:35:11.925 align:middle line:90% Yeah, question? 00:35:11.925 --> 00:35:12.550 align:middle line:90% AUDIENCE: Yeah. 00:35:12.550 --> 00:35:15.870 align:middle line:84% So for a person that moved from Georgia to Florida, for example, 00:35:15.870 --> 00:35:18.450 align:middle line:84% because there's no state income tax in Florida, 00:35:18.450 --> 00:35:21.450 align:middle line:84% then for those people, they weren't in this equilibrium 00:35:21.450 --> 00:35:22.350 align:middle line:90% in Georgia, right? 00:35:22.350 --> 00:35:22.930 align:middle line:90% [INAUDIBLE] 00:35:22.930 --> 00:35:24.930 align:middle line:84% JON GRUBER: I mean, if that's the reason they moved. 00:35:24.930 --> 00:35:26.510 align:middle line:84% Once again, they might have moved for other reasons. 00:35:26.510 --> 00:35:28.552 align:middle line:84% But if they're voting with their feet and saying, 00:35:28.552 --> 00:35:31.130 align:middle line:84% look, Georgia's got all these taxes-- 00:35:31.130 --> 00:35:33.570 align:middle line:84% why do people move to Florida when they're retired? 00:35:33.570 --> 00:35:35.470 align:middle line:84% No state income tax because it's warm, 00:35:35.470 --> 00:35:36.910 align:middle line:90% but also no state income tax. 00:35:36.910 --> 00:35:39.470 align:middle line:84% So why should I pay taxes to support schools? 00:35:39.470 --> 00:35:40.510 align:middle line:90% Screw that. 00:35:40.510 --> 00:35:42.255 align:middle line:90% I'm a selfish old person. 00:35:42.255 --> 00:35:44.130 align:middle line:84% I'm moving to Florida, and I won't pay taxes. 00:35:44.130 --> 00:35:45.463 align:middle line:90% And the schools suck in Florida. 00:35:45.463 --> 00:35:46.890 align:middle line:90% No offense, but they do. 00:35:46.890 --> 00:35:49.910 align:middle line:90% 00:35:49.910 --> 00:35:51.443 align:middle line:90% And so that's a Tiebout outcome. 00:35:51.443 --> 00:35:53.610 align:middle line:84% People who don't care about schools move to Florida. 00:35:53.610 --> 00:35:55.450 align:middle line:84% People who do care about schools move to places 00:35:55.450 --> 00:35:56.430 align:middle line:90% where schools are good. 00:35:56.430 --> 00:36:00.370 align:middle line:90% 00:36:00.370 --> 00:36:04.010 align:middle line:90% So that says that basically-- 00:36:04.010 --> 00:36:06.070 align:middle line:84% well, let me go back to one thing. 00:36:06.070 --> 00:36:08.690 align:middle line:84% What is the actual evidence that capitalization exists? 00:36:08.690 --> 00:36:12.110 align:middle line:84% And here, we have a wonderful example from one 00:36:12.110 --> 00:36:16.430 align:middle line:84% of the most important events in our nation's history 00:36:16.430 --> 00:36:22.910 align:middle line:84% of fiscal federalism, which was California's Proposition 13. 00:36:22.910 --> 00:36:29.390 align:middle line:84% Proposition 13 was a law passed by California in 1978 00:36:29.390 --> 00:36:34.250 align:middle line:84% that limited local towns ability to raise property taxes. 00:36:34.250 --> 00:36:36.590 align:middle line:84% Before this law, all over the country, 00:36:36.590 --> 00:36:38.310 align:middle line:84% towns could do whatever they wanted. 00:36:38.310 --> 00:36:40.730 align:middle line:84% California was the first state to say, wait a second, 00:36:40.730 --> 00:36:43.690 align:middle line:84% we are not going to let towns set their own property tax. 00:36:43.690 --> 00:36:44.790 align:middle line:90% We're going to limit them. 00:36:44.790 --> 00:36:47.350 align:middle line:84% In particular, what the law said is 00:36:47.350 --> 00:36:49.830 align:middle line:84% that you could not have a property 00:36:49.830 --> 00:36:52.910 align:middle line:84% tax that was more than 1% of a home's assessed value. 00:36:52.910 --> 00:36:54.962 align:middle line:84% So every year people could not pay more than 1%. 00:36:54.962 --> 00:36:56.670 align:middle line:84% Just to fix ideas, property tax in the US 00:36:56.670 --> 00:37:01.770 align:middle line:84% are typically between 1 and 2.5% or between half a percent, 2.5%. 00:37:01.770 --> 00:37:04.170 align:middle line:84% But we'll cover property taxes in chapter 23. 00:37:04.170 --> 00:37:05.690 align:middle line:90% So that's a pretty strict limit. 00:37:05.690 --> 00:37:09.850 align:middle line:84% Moreover, it could not be more than 1% of the home's value 00:37:09.850 --> 00:37:11.290 align:middle line:90% at that point in time. 00:37:11.290 --> 00:37:13.450 align:middle line:84% And it could only increase at 2% per year 00:37:13.450 --> 00:37:17.090 align:middle line:84% no matter what happened to the value of the house. 00:37:17.090 --> 00:37:22.810 align:middle line:90% So think about what that means. 00:37:22.810 --> 00:37:25.910 align:middle line:84% So a great study in 1982 looked at the effect of Proposition 13. 00:37:25.910 --> 00:37:27.890 align:middle line:84% And he said, look, let's compare towns 00:37:27.890 --> 00:37:32.450 align:middle line:84% that before this law already had low property taxes to ones 00:37:32.450 --> 00:37:34.450 align:middle line:90% that had high property taxes. 00:37:34.450 --> 00:37:36.210 align:middle line:84% Those high property taxes had to reduce 00:37:36.210 --> 00:37:38.290 align:middle line:90% their property taxes a ton. 00:37:38.290 --> 00:37:40.570 align:middle line:84% The ones that low property taxes didn't. 00:37:40.570 --> 00:37:43.450 align:middle line:84% So let's ask what happened to house values. 00:37:43.450 --> 00:37:46.170 align:middle line:84% And what he found was an enormous increase in house 00:37:46.170 --> 00:37:49.510 align:middle line:84% values, an enormous increase in house values. 00:37:49.510 --> 00:37:55.510 align:middle line:84% Indeed, what he found was that basically, 00:37:55.510 --> 00:37:58.730 align:middle line:84% for every $1 in property tax reduction, 00:37:58.730 --> 00:38:02.380 align:middle line:84% for every $1 per year in property tax reduction, 00:38:02.380 --> 00:38:06.700 align:middle line:84% the value of homes in California went up by $7. 00:38:06.700 --> 00:38:10.040 align:middle line:84% Now your reaction is, wait a second, how can that be so big? 00:38:10.040 --> 00:38:11.280 align:middle line:90% But why does that make sense? 00:38:11.280 --> 00:38:12.655 align:middle line:84% Why does it make sense that you'd 00:38:12.655 --> 00:38:16.100 align:middle line:84% get $1 per year leading to $7 in house value? 00:38:16.100 --> 00:38:17.080 align:middle line:90% Yeah? 00:38:17.080 --> 00:38:18.500 align:middle line:84% AUDIENCE: Because you hold the house for a long time. 00:38:18.500 --> 00:38:19.720 align:middle line:84% JON GRUBER: Because You hold the house for a long time. 00:38:19.720 --> 00:38:22.700 align:middle line:84% Indeed if we think you hold the house forever, 00:38:22.700 --> 00:38:28.560 align:middle line:84% what is $7 worth in [? PD? ?] Basically, what is $7? 00:38:28.560 --> 00:38:29.720 align:middle line:90% What should it compare to? 00:38:29.720 --> 00:38:31.280 align:middle line:90% Well, you're paying $1 a year. 00:38:31.280 --> 00:38:32.040 align:middle line:90% You've been paying it forever. 00:38:32.040 --> 00:38:33.560 align:middle line:84% You put it over the interest rate. 00:38:33.560 --> 00:38:35.977 align:middle line:84% When this law passed, interest rates were very, very high. 00:38:35.977 --> 00:38:38.900 align:middle line:84% Interest rates were typically about 12%. 00:38:38.900 --> 00:38:42.420 align:middle line:90% So this equals $8.33. 00:38:42.420 --> 00:38:46.925 align:middle line:84% So, actually, the capitalization was pretty close to 100%, 00:38:46.925 --> 00:38:49.300 align:middle line:84% pretty close to if you look at the PDV of what people are 00:38:49.300 --> 00:38:50.980 align:middle line:90% going to owe in property taxes. 00:38:50.980 --> 00:38:52.780 align:middle line:84% When you lower taxes, house prices 00:38:52.780 --> 00:38:55.460 align:middle line:84% went up just about dollar for dollar. 00:38:55.460 --> 00:38:57.740 align:middle line:90% So what does this imply? 00:38:57.740 --> 00:39:02.440 align:middle line:84% This implies that people thought we were massively overproviding 00:39:02.440 --> 00:39:08.240 align:middle line:84% public goods, or they thought the state would bail them out. 00:39:08.240 --> 00:39:10.160 align:middle line:84% And maybe people thought, yeah, we'll do this, 00:39:10.160 --> 00:39:12.000 align:middle line:84% and the state will realize our schools are going to fall apart 00:39:12.000 --> 00:39:13.180 align:middle line:90% and they'll send us money. 00:39:13.180 --> 00:39:16.560 align:middle line:84% So this is just the way we squeeze money out of the state. 00:39:16.560 --> 00:39:21.000 align:middle line:84% OK, well, it turned out they were wrong. 00:39:21.000 --> 00:39:24.560 align:middle line:84% California went from having top 10 school system in the nation 00:39:24.560 --> 00:39:26.460 align:middle line:84% to bottom 10 school system in the nation. 00:39:26.460 --> 00:39:28.480 align:middle line:84% California public schools went to shit. 00:39:28.480 --> 00:39:30.640 align:middle line:84% Everybody left and went to private school. 00:39:30.640 --> 00:39:33.980 align:middle line:84% And basically, the California public school system 00:39:33.980 --> 00:39:36.380 align:middle line:90% has suffered massively. 00:39:36.380 --> 00:39:40.840 align:middle line:90% 00:39:40.840 --> 00:39:45.320 align:middle line:84% So basically, we have a system where essentially people-- 00:39:45.320 --> 00:39:49.008 align:middle line:84% now, maybe that's what people wanted. 00:39:49.008 --> 00:39:51.300 align:middle line:84% Maybe people wanted crappy public schools at low price. 00:39:51.300 --> 00:39:52.540 align:middle line:90% Nothing wrong with that. 00:39:52.540 --> 00:39:54.540 align:middle line:84% The question is is it really what they wanted? 00:39:54.540 --> 00:39:57.340 align:middle line:84% Did they really understand that that was a consequence? 00:39:57.340 --> 00:39:58.840 align:middle line:84% Or do they think maybe, nah, they'll 00:39:58.840 --> 00:40:00.100 align:middle line:84% figure some other way to pay for the schools? 00:40:00.100 --> 00:40:01.280 align:middle line:90% Maybe they were misinformed. 00:40:01.280 --> 00:40:03.322 align:middle line:84% Maybe they thought the state would bail them out. 00:40:03.322 --> 00:40:05.220 align:middle line:90% 00:40:05.220 --> 00:40:08.500 align:middle line:84% Indeed, more recent work suggests 00:40:08.500 --> 00:40:12.060 align:middle line:84% that on average in the US, more spending on schools 00:40:12.060 --> 00:40:14.540 align:middle line:84% would actually raise house values. 00:40:14.540 --> 00:40:16.860 align:middle line:84% More spending on schools financed by property tax 00:40:16.860 --> 00:40:17.920 align:middle line:90% would raise house values. 00:40:17.920 --> 00:40:20.540 align:middle line:84% There's a cool study which looked 00:40:20.540 --> 00:40:24.340 align:middle line:84% at what happens when towns hold special elections to raise 00:40:24.340 --> 00:40:26.940 align:middle line:84% certain supplemental funds to fund capital 00:40:26.940 --> 00:40:28.260 align:middle line:90% improvements in schools. 00:40:28.260 --> 00:40:29.885 align:middle line:84% So a town will have a special election. 00:40:29.885 --> 00:40:32.260 align:middle line:84% Everybody needs to pay $500 more or another 0.1% 00:40:32.260 --> 00:40:33.200 align:middle line:90% in the property tax. 00:40:33.200 --> 00:40:35.180 align:middle line:84% And we're going to rebuild all the schools. 00:40:35.180 --> 00:40:38.540 align:middle line:84% They found when they do that, house values go up, 00:40:38.540 --> 00:40:41.340 align:middle line:84% not down as a California example would suggest. 00:40:41.340 --> 00:40:45.500 align:middle line:84% So higher taxes, funding better schools make it go up. 00:40:45.500 --> 00:40:49.897 align:middle line:84% So what might be going on in California too was just, 00:40:49.897 --> 00:40:51.480 align:middle line:84% OK, we don't think the schools suffer. 00:40:51.480 --> 00:40:53.640 align:middle line:84% We think politicians are wasting a lot of money on other stuff, 00:40:53.640 --> 00:40:55.973 align:middle line:84% and that will all go away, and our schools won't suffer. 00:40:55.973 --> 00:40:57.300 align:middle line:90% And folks were just wrong. 00:40:57.300 --> 00:40:59.838 align:middle line:84% And now California has crappy public schools. 00:40:59.838 --> 00:41:02.380 align:middle line:84% So it's an interesting example of capitalization in practice. 00:41:02.380 --> 00:41:04.120 align:middle line:84% The capitalization really worked. 00:41:04.120 --> 00:41:06.640 align:middle line:84% The question is is it really a test of Tiebout? 00:41:06.640 --> 00:41:10.960 align:middle line:84% That depends on this, information friction. 00:41:10.960 --> 00:41:14.812 align:middle line:84% Did people really understand what they were voting for? 00:41:14.812 --> 00:41:17.020 align:middle line:84% Because at least later evidence suggests they didn't. 00:41:17.020 --> 00:41:17.960 align:middle line:84% Because when people have the choice 00:41:17.960 --> 00:41:20.560 align:middle line:84% to just vote about what I pay more for better schools, 00:41:20.560 --> 00:41:22.800 align:middle line:84% typically, when they do so, that's 00:41:22.800 --> 00:41:25.520 align:middle line:90% better for their housing value. 00:41:25.520 --> 00:41:26.440 align:middle line:90% Questions about that? 00:41:26.440 --> 00:41:29.200 align:middle line:90% 00:41:29.200 --> 00:41:31.920 align:middle line:84% Where does this leave us in the question we started with, 00:41:31.920 --> 00:41:34.340 align:middle line:84% which is optimal fiscal federalism? 00:41:34.340 --> 00:41:36.920 align:middle line:90% 00:41:36.920 --> 00:41:38.700 align:middle line:90% Optimal fiscal federalism. 00:41:38.700 --> 00:41:41.120 align:middle line:84% Where does this leave us in optimal fiscal federalism? 00:41:41.120 --> 00:41:42.880 align:middle line:90% Where it leaves us is-- 00:41:42.880 --> 00:41:50.880 align:middle line:84% basically, the Tiebout model says that government spending 00:41:50.880 --> 00:41:54.800 align:middle line:84% is most efficient where the tax benefit linkages are 00:41:54.800 --> 00:41:59.540 align:middle line:84% the largest, that where the tax benefit linkages are 00:41:59.540 --> 00:42:02.723 align:middle line:84% the largest, you have the least inefficiency because people 00:42:02.723 --> 00:42:05.140 align:middle line:84% can vote with their feet and get the level of public goods 00:42:05.140 --> 00:42:07.140 align:middle line:90% they want. 00:42:07.140 --> 00:42:12.700 align:middle line:90% So basically, that says-- 00:42:12.700 --> 00:42:15.020 align:middle line:84% so what the Tiebout model says is optimal 00:42:15.020 --> 00:42:17.220 align:middle line:84% fiscal federalism is determined by three things. 00:42:17.220 --> 00:42:21.420 align:middle line:90% One is tax-benefit links. 00:42:21.420 --> 00:42:23.740 align:middle line:84% Two, the second thing is determined by 00:42:23.740 --> 00:42:31.060 align:middle line:84% is the level of spillovers because the more spillovers, 00:42:31.060 --> 00:42:33.620 align:middle line:84% the more that it'll be inefficient to do it locally. 00:42:33.620 --> 00:42:35.620 align:middle line:84% And the third thing is going to be determined by 00:42:35.620 --> 00:42:37.860 align:middle line:90% is economies of scale. 00:42:37.860 --> 00:42:41.300 align:middle line:84% Because basically, something which has to be huge to succeed 00:42:41.300 --> 00:42:43.100 align:middle line:90% shouldn't be provided locally. 00:42:43.100 --> 00:42:45.518 align:middle line:84% So in other words, if we go back to chapter 1 00:42:45.518 --> 00:42:48.060 align:middle line:84% and that mystery of some stuff was done federally, some stuff 00:42:48.060 --> 00:42:52.300 align:middle line:84% was done locally, we can ask, do these three factors explain it? 00:42:52.300 --> 00:42:55.360 align:middle line:90% And the answer is they do. 00:42:55.360 --> 00:42:56.980 align:middle line:84% What does the federal government do? 00:42:56.980 --> 00:43:00.240 align:middle line:84% Defense, huge economies of scale. 00:43:00.240 --> 00:43:03.400 align:middle line:84% Very hard to make a tax-benefit linkage on defense. 00:43:03.400 --> 00:43:06.440 align:middle line:90% OK, what do local economies do? 00:43:06.440 --> 00:43:07.887 align:middle line:90% Fix the roads. 00:43:07.887 --> 00:43:09.220 align:middle line:90% Not that big a economy of scale. 00:43:09.220 --> 00:43:10.960 align:middle line:84% You just need a guy in a machine. 00:43:10.960 --> 00:43:12.820 align:middle line:84% OK, but a lot of tax-benefit linkage. 00:43:12.820 --> 00:43:15.600 align:middle line:84% I can see that my road got better when I pay higher taxes. 00:43:15.600 --> 00:43:19.220 align:middle line:84% Schools, some economies of scale, but a lot of linkages. 00:43:19.220 --> 00:43:22.040 align:middle line:90% Police, same thing. 00:43:22.040 --> 00:43:24.780 align:middle line:84% So, actually the division of responsibilities makes sense. 00:43:24.780 --> 00:43:26.440 align:middle line:84% The main thing you have from the Tiebout framework 00:43:26.440 --> 00:43:28.500 align:middle line:84% is stuff's kind of divided the way it should be. 00:43:28.500 --> 00:43:31.720 align:middle line:84% Think about redistribution, especially social insurance 00:43:31.720 --> 00:43:32.620 align:middle line:90% programs. 00:43:32.620 --> 00:43:35.140 align:middle line:84% They're done at the state and mostly federal level. 00:43:35.140 --> 00:43:35.680 align:middle line:90% Why? 00:43:35.680 --> 00:43:38.640 align:middle line:84% Why don't we have local redistribution? 00:43:38.640 --> 00:43:41.920 align:middle line:84% AUDIENCE: Because people tend to locally be separated 00:43:41.920 --> 00:43:43.720 align:middle line:90% by socioeconomic status. 00:43:43.720 --> 00:43:45.100 align:middle line:90% JON GRUBER: Well, that's right. 00:43:45.100 --> 00:43:45.815 align:middle line:90% But why else? 00:43:45.815 --> 00:43:48.440 align:middle line:84% I mean, they do, but it's still not fully-- why don't we just-- 00:43:48.440 --> 00:43:49.400 align:middle line:84% in Lexington, why don't we say just 00:43:49.400 --> 00:43:51.520 align:middle line:84% get rid of federal income tax, have a big Lexington income tax 00:43:51.520 --> 00:43:53.797 align:middle line:84% and distribute to the poor people of Lexington? 00:43:53.797 --> 00:43:54.630 align:middle line:90% Why won't that work? 00:43:54.630 --> 00:43:58.070 align:middle line:90% What does that violate here? 00:43:58.070 --> 00:43:58.570 align:middle line:90% Yeah? 00:43:58.570 --> 00:44:00.270 align:middle line:90% AUDIENCE: Spillover. 00:44:00.270 --> 00:44:00.990 align:middle line:90% JON GRUBER: No. 00:44:00.990 --> 00:44:01.490 align:middle line:90% Yeah? 00:44:01.490 --> 00:44:03.825 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:44:03.825 --> 00:44:05.010 align:middle line:90% JON GRUBER: Yeah. 00:44:05.010 --> 00:44:07.650 align:middle line:84% Basically, if Lexington starts taxing all the rich people 00:44:07.650 --> 00:44:09.110 align:middle line:84% to give to the poor people, they'll leave. 00:44:09.110 --> 00:44:10.943 align:middle line:84% If America does, they're not going to leave. 00:44:10.943 --> 00:44:14.250 align:middle line:84% I mean, some will, they'll go to Macao or whatever, or no 00:44:14.250 --> 00:44:15.570 align:middle line:90% Monaco or whatever. 00:44:15.570 --> 00:44:19.890 align:middle line:84% But basically, the point is the tax-benefit linkages 00:44:19.890 --> 00:44:23.690 align:middle line:84% are sufficiently weak that if a town tries to do it, 00:44:23.690 --> 00:44:25.650 align:middle line:90% they'll get killed. 00:44:25.650 --> 00:44:28.610 align:middle line:84% So basically, that's why redistribution is mostly 00:44:28.610 --> 00:44:30.970 align:middle line:84% done at higher levels of government 00:44:30.970 --> 00:44:32.550 align:middle line:90% is for tax-benefit linkages. 00:44:32.550 --> 00:44:34.842 align:middle line:84% Things like defense are at a higher level of government 00:44:34.842 --> 00:44:36.610 align:middle line:90% because of economies of scale. 00:44:36.610 --> 00:44:39.073 align:middle line:84% Things like roads and police are done 00:44:39.073 --> 00:44:41.490 align:middle line:84% at lower levels of government because the economy of scale 00:44:41.490 --> 00:44:44.370 align:middle line:84% is large and the tax-benefit linkages are tight. 00:44:44.370 --> 00:44:45.170 align:middle line:90% Yeah? 00:44:45.170 --> 00:44:46.990 align:middle line:84% AUDIENCE: What's an example of spillovers? 00:44:46.990 --> 00:44:50.970 align:middle line:84% JON GRUBER: OK, well, let's come to the one category-- 00:44:50.970 --> 00:44:54.750 align:middle line:84% I would argue I'm pretty much every category we have, 00:44:54.750 --> 00:44:58.280 align:middle line:84% the US gets it right, except for one, 00:44:58.280 --> 00:45:00.030 align:middle line:84% where we differ from the rest of the world 00:45:00.030 --> 00:45:04.030 align:middle line:84% and we probably get it somewhat wrong, which is education. 00:45:04.030 --> 00:45:07.710 align:middle line:84% Education, as we'll talk about next chapter, is about 50% 00:45:07.710 --> 00:45:09.110 align:middle line:90% locally financed. 00:45:09.110 --> 00:45:13.830 align:middle line:84% It's about 45% locally financed, 45% state financed, 00:45:13.830 --> 00:45:16.630 align:middle line:84% 10% federally financed or probably even less. 00:45:16.630 --> 00:45:20.590 align:middle line:84% The US federal government does very little education. 00:45:20.590 --> 00:45:21.810 align:middle line:90% Why could that be wrong? 00:45:21.810 --> 00:45:23.268 align:middle line:84% And why do other countries actually 00:45:23.268 --> 00:45:27.630 align:middle line:84% have the central government play a much larger role in education? 00:45:27.630 --> 00:45:30.750 align:middle line:84% And the example is spillovers, that basically when 00:45:30.750 --> 00:45:35.270 align:middle line:84% I educate someone in my town, that benefits the whole country. 00:45:35.270 --> 00:45:36.953 align:middle line:90% They pay higher federal taxes. 00:45:36.953 --> 00:45:38.870 align:middle line:84% They might start a company in some other place 00:45:38.870 --> 00:45:40.330 align:middle line:90% which hires lots of people. 00:45:40.330 --> 00:45:44.265 align:middle line:84% Now, it does benefit a lot-- most people don't move. 00:45:44.265 --> 00:45:45.390 align:middle line:90% I love a great line, I had. 00:45:45.390 --> 00:45:46.493 align:middle line:90% Everybody talks about-- 00:45:46.493 --> 00:45:48.410 align:middle line:84% I like to think I married the love of my life. 00:45:48.410 --> 00:45:49.730 align:middle line:84% Everybody says they married the love of their life. 00:45:49.730 --> 00:45:51.730 align:middle line:84% They said it is interesting that most people happen 00:45:51.730 --> 00:45:54.313 align:middle line:84% to have the love of their life living within 10 miles of where 00:45:54.313 --> 00:45:57.430 align:middle line:84% they were born because most people don't move. 00:45:57.430 --> 00:46:00.010 align:middle line:84% So it is true that most of the benefits 00:46:00.010 --> 00:46:02.650 align:middle line:84% accrue to states and localities, but it's 00:46:02.650 --> 00:46:04.650 align:middle line:84% hard to argue less than 10% of the benefits 00:46:04.650 --> 00:46:08.210 align:middle line:84% accrue to the country of having a better educated population. 00:46:08.210 --> 00:46:10.070 align:middle line:84% So for that, that's a spillover argument. 00:46:10.070 --> 00:46:12.370 align:middle line:84% And that's example of why most other countries actually 00:46:12.370 --> 00:46:15.030 align:middle line:84% finance education much more to central level than we do. 00:46:15.030 --> 00:46:18.610 align:middle line:90% And we probably should. 00:46:18.610 --> 00:46:20.250 align:middle line:90% Questions about that? 00:46:20.250 --> 00:46:20.827 align:middle line:90% Yeah? 00:46:20.827 --> 00:46:23.410 align:middle line:84% AUDIENCE: Would health care be another area with big spillover 00:46:23.410 --> 00:46:25.302 align:middle line:90% effects? 00:46:25.302 --> 00:46:27.310 align:middle line:90% JON GRUBER: Health care could. 00:46:27.310 --> 00:46:28.062 align:middle line:90% Yes, absolutely. 00:46:28.062 --> 00:46:29.270 align:middle line:90% Health care would be another. 00:46:29.270 --> 00:46:31.687 align:middle line:84% And that's once again not really state, 00:46:31.687 --> 00:46:32.770 align:middle line:90% but that one we get right. 00:46:32.770 --> 00:46:35.062 align:middle line:84% That's federally and state financed, not really locally 00:46:35.062 --> 00:46:37.250 align:middle line:90% financed. 00:46:37.250 --> 00:46:39.430 align:middle line:84% Once again, these theories are kind of wacky. 00:46:39.430 --> 00:46:41.930 align:middle line:84% But, actually, they explain a lot of central patterns 00:46:41.930 --> 00:46:42.870 align:middle line:90% pretty well. 00:46:42.870 --> 00:46:44.490 align:middle line:84% Like these three factors explain a lot 00:46:44.490 --> 00:46:46.740 align:middle line:84% of why stuff is done at different levels of government 00:46:46.740 --> 00:46:48.390 align:middle line:90% pretty well. 00:46:48.390 --> 00:46:50.370 align:middle line:90% Other questions or comments? 00:46:50.370 --> 00:46:50.870 align:middle line:90% Yeah? 00:46:50.870 --> 00:46:51.630 align:middle line:90% AUDIENCE: At what level-- 00:46:51.630 --> 00:46:53.230 align:middle line:84% I'm thinking like a township might 00:46:53.230 --> 00:46:55.930 align:middle line:84% want to do something versus the state versus the United States. 00:46:55.930 --> 00:46:57.950 align:middle line:84% Like I agree that people aren't going to move. 00:46:57.950 --> 00:46:59.850 align:middle line:84% Most of them aren't going to leave America. 00:46:59.850 --> 00:47:02.650 align:middle line:84% It's like America starts like doing redistribution. 00:47:02.650 --> 00:47:06.250 align:middle line:84% I guess this is more politically oriented, but maybe not. 00:47:06.250 --> 00:47:09.430 align:middle line:84% I guess the question would be, at what scale does 00:47:09.430 --> 00:47:13.188 align:middle line:84% the tax-benefit links outweigh the-- 00:47:13.188 --> 00:47:14.730 align:middle line:84% JON GRUBER: I mean, it's a trade off. 00:47:14.730 --> 00:47:15.410 align:middle line:90% We don't know. 00:47:15.410 --> 00:47:17.330 align:middle line:84% I mean, obviously, like for example, 00:47:17.330 --> 00:47:19.930 align:middle line:84% one of the main things that localities do is roads. 00:47:19.930 --> 00:47:25.070 align:middle line:84% There are spillovers with roads because you drive through. 00:47:25.070 --> 00:47:29.630 align:middle line:84% There are economies of scale with roads. 00:47:29.630 --> 00:47:32.990 align:middle line:84% It's more the forces in each direction, the forces that 00:47:32.990 --> 00:47:34.090 align:middle line:90% push in each direction. 00:47:34.090 --> 00:47:34.590 align:middle line:90% Yeah? 00:47:34.590 --> 00:47:35.898 align:middle line:90% AUDIENCE: OK, so [INAUDIBLE]. 00:47:35.898 --> 00:47:39.870 align:middle line:90% 00:47:39.870 --> 00:47:44.820 align:middle line:84% Like on a local level it would be like [INAUDIBLE] 00:47:44.820 --> 00:47:46.070 align:middle line:90% JON GRUBER: Which public good? 00:47:46.070 --> 00:47:48.582 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:47:48.582 --> 00:47:50.290 align:middle line:84% JON GRUBER: Like if you do redistribution 00:47:50.290 --> 00:47:52.695 align:middle line:84% at the local level, that would be underprovided 00:47:52.695 --> 00:47:55.070 align:middle line:84% because the problem is if you try to distribute too much, 00:47:55.070 --> 00:47:56.513 align:middle line:90% everybody leaves your town. 00:47:56.513 --> 00:47:57.930 align:middle line:84% So you end up in equilibrium where 00:47:57.930 --> 00:47:59.930 align:middle line:90% you can't do redistribution. 00:47:59.930 --> 00:48:07.250 align:middle line:84% AUDIENCE: Well, my question is more related to [INAUDIBLE] 00:48:07.250 --> 00:48:10.230 align:middle line:84% if this should be done at a federal level? 00:48:10.230 --> 00:48:11.050 align:middle line:90% JON GRUBER: Yes. 00:48:11.050 --> 00:48:14.290 align:middle line:84% AUDIENCE: [INAUDIBLE] provision of a public good that 00:48:14.290 --> 00:48:14.970 align:middle line:90% [INAUDIBLE] 00:48:14.970 --> 00:48:16.950 align:middle line:84% JON GRUBER: So basically, here's another way to put it. 00:48:16.950 --> 00:48:18.075 align:middle line:90% It's a great way to put it. 00:48:18.075 --> 00:48:21.570 align:middle line:84% More tax benefit linkages means more local. 00:48:21.570 --> 00:48:24.170 align:middle line:84% More spillovers means less local. 00:48:24.170 --> 00:48:26.535 align:middle line:84% More economies of scale means less local. 00:48:26.535 --> 00:48:27.910 align:middle line:84% That's the way to think about it. 00:48:27.910 --> 00:48:29.310 align:middle line:90% Those are the forces pushing. 00:48:29.310 --> 00:48:31.470 align:middle line:84% That's a good way to answer the question, too. 00:48:31.470 --> 00:48:32.610 align:middle line:90% Yeah? 00:48:32.610 --> 00:48:37.090 align:middle line:84% AUDIENCE: How relevant is this [INAUDIBLE] immigration 00:48:37.090 --> 00:48:38.893 align:middle line:84% or migration into [INAUDIBLE] country? 00:48:38.893 --> 00:48:39.810 align:middle line:90% JON GRUBER: Excellent. 00:48:39.810 --> 00:48:43.170 align:middle line:84% I mean, look, we will talk-- it's not that there's 00:48:43.170 --> 00:48:44.750 align:middle line:90% no mobility across countries. 00:48:44.750 --> 00:48:48.360 align:middle line:84% There's obviously mobility across countries. 00:48:48.360 --> 00:48:50.237 align:middle line:84% But I think there is not a lot of evidence 00:48:50.237 --> 00:48:51.820 align:middle line:84% that mobility for most people responds 00:48:51.820 --> 00:48:53.160 align:middle line:90% to local public goods or taxes. 00:48:53.160 --> 00:48:55.482 align:middle line:84% It's more about not getting killed. 00:48:55.482 --> 00:48:56.940 align:middle line:84% That is not true, however, totally. 00:48:56.940 --> 00:48:58.398 align:middle line:84% There's a lot of cool studies we'll 00:48:58.398 --> 00:49:00.220 align:middle line:84% talk about when we get to chapter 25 00:49:00.220 --> 00:49:03.020 align:middle line:84% about very rich people moving across countries in response 00:49:03.020 --> 00:49:06.860 align:middle line:84% to the top marginal tax rate, particularly athletes. 00:49:06.860 --> 00:49:09.380 align:middle line:84% But it's generally not really true. 00:49:09.380 --> 00:49:12.300 align:middle line:90% OK, other questions or comments? 00:49:12.300 --> 00:49:16.440 align:middle line:84% All right, now that's all optimal fiscal federalism. 00:49:16.440 --> 00:49:18.500 align:middle line:84% Now having done that, we now come to the part 00:49:18.500 --> 00:49:21.420 align:middle line:84% that we really care about, which is, OK, we 00:49:21.420 --> 00:49:25.340 align:middle line:84% know what should be optimal, but what should we 00:49:25.340 --> 00:49:28.300 align:middle line:84% do to try to encourage the optimal level 00:49:28.300 --> 00:49:29.980 align:middle line:84% of fiscal federalism, the optimal level 00:49:29.980 --> 00:49:31.940 align:middle line:90% of public provision? 00:49:31.940 --> 00:49:34.740 align:middle line:84% And now we get to the last topic, 00:49:34.740 --> 00:49:38.140 align:middle line:84% which is the role of intergovernmental grants 00:49:38.140 --> 00:49:40.820 align:middle line:84% and thinking about how government should behave 00:49:40.820 --> 00:49:43.340 align:middle line:90% to get at that optimal level. 00:49:43.340 --> 00:49:45.708 align:middle line:84% So the first thing is let's step back 00:49:45.708 --> 00:49:48.000 align:middle line:84% and just talk about the general topic of redistribution 00:49:48.000 --> 00:49:49.953 align:middle line:90% across communities. 00:49:49.953 --> 00:49:52.120 align:middle line:84% Just think about the general topic of redistributing 00:49:52.120 --> 00:49:53.800 align:middle line:90% across communities. 00:49:53.800 --> 00:49:55.720 align:middle line:84% So let me ask you a general question. 00:49:55.720 --> 00:49:58.560 align:middle line:84% Given all you've learned, should we 00:49:58.560 --> 00:50:02.800 align:middle line:84% or should we not redistribute from, say, rich communities 00:50:02.800 --> 00:50:03.900 align:middle line:90% to poor communities? 00:50:03.900 --> 00:50:07.605 align:middle line:90% 00:50:07.605 --> 00:50:08.980 align:middle line:84% I'm not talking about rich people 00:50:08.980 --> 00:50:12.060 align:middle line:84% to poor people, rich communities to poor communities. 00:50:12.060 --> 00:50:14.280 align:middle line:84% Should we, in general, take money from rich places 00:50:14.280 --> 00:50:16.200 align:middle line:90% and give it to poor places? 00:50:16.200 --> 00:50:18.680 align:middle line:84% Should we or should we not do that? 00:50:18.680 --> 00:50:20.520 align:middle line:84% The answer's, of course, ambiguous. 00:50:20.520 --> 00:50:22.380 align:middle line:84% What are the factors pushing each direction? 00:50:22.380 --> 00:50:25.460 align:middle line:90% 00:50:25.460 --> 00:50:25.960 align:middle line:90% Yeah? 00:50:25.960 --> 00:50:28.000 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:50:28.000 --> 00:50:29.000 align:middle line:90% JON GRUBER: What's that? 00:50:29.000 --> 00:50:31.400 align:middle line:84% AUDIENCE: There are a lot of [INAUDIBLE] things 00:50:31.400 --> 00:50:33.720 align:middle line:84% that you can achieve in a higher level 00:50:33.720 --> 00:50:42.420 align:middle line:84% if you redistribute that [INAUDIBLE] 00:50:42.420 --> 00:50:44.192 align:middle line:84% JON GRUBER: So in some sense a reason-- 00:50:44.192 --> 00:50:45.900 align:middle line:84% so let me just shortcut this a little bit 00:50:45.900 --> 00:50:47.733 align:middle line:84% because we're getting a little late on time. 00:50:47.733 --> 00:50:50.140 align:middle line:84% The reason not to distribute is Tiebout. 00:50:50.140 --> 00:50:53.300 align:middle line:84% The bottom line is if Tiebout holds, 00:50:53.300 --> 00:50:55.740 align:middle line:90% you don't want to redistribute. 00:50:55.740 --> 00:50:58.060 align:middle line:84% You might want distribute from rich to poor people, 00:50:58.060 --> 00:51:00.060 align:middle line:84% but there's no reason to take from one community 00:51:00.060 --> 00:51:01.980 align:middle line:84% to give to another if everybody's happy. 00:51:01.980 --> 00:51:04.060 align:middle line:84% In a Tiebout world, everybody's happy. 00:51:04.060 --> 00:51:06.820 align:middle line:84% It can't be welfare improving to redistribute across communities 00:51:06.820 --> 00:51:08.820 align:middle line:84% in a Tiebout world because everyone's 00:51:08.820 --> 00:51:11.200 align:middle line:84% chosen their optimal level of taxes and public goods. 00:51:11.200 --> 00:51:13.140 align:middle line:84% I'm not saying you shouldn't have a federal income tax when 00:51:13.140 --> 00:51:14.420 align:middle line:90% you switch from rich to poor. 00:51:14.420 --> 00:51:16.587 align:middle line:84% But there's no reason to actually take from one town 00:51:16.587 --> 00:51:18.500 align:middle line:84% and give-- just because my town has 00:51:18.500 --> 00:51:20.740 align:middle line:84% rich people with good schools and another town 00:51:20.740 --> 00:51:22.980 align:middle line:84% has poor people with bad schools, Tiebout would say, 00:51:22.980 --> 00:51:24.660 align:middle line:90% that's the choice they made. 00:51:24.660 --> 00:51:26.297 align:middle line:90% There's no reason to distribute. 00:51:26.297 --> 00:51:27.880 align:middle line:84% If you care about income distribution, 00:51:27.880 --> 00:51:29.280 align:middle line:84% you do have a federal income tax. 00:51:29.280 --> 00:51:30.863 align:middle line:84% There's no reason to take from my town 00:51:30.863 --> 00:51:32.820 align:middle line:90% and give it to the other town. 00:51:32.820 --> 00:51:34.400 align:middle line:90% Why is Tiebout wrong? 00:51:34.400 --> 00:51:35.400 align:middle line:90% Where he wrong? 00:51:35.400 --> 00:51:39.783 align:middle line:84% He's wrong if people can't freely move. 00:51:39.783 --> 00:51:41.700 align:middle line:84% In other words, if redistribution does matter, 00:51:41.700 --> 00:51:43.840 align:middle line:90% then Tiebout doesn't hold. 00:51:43.840 --> 00:51:45.960 align:middle line:84% So essentially it's a simple equation, 00:51:45.960 --> 00:51:49.040 align:middle line:84% which is the more Tiebout holds, the less you 00:51:49.040 --> 00:51:52.600 align:middle line:84% want to be doing community-level redistribution. 00:51:52.600 --> 00:51:55.780 align:middle line:84% The less Tiebout holds, the more you just want to be considering. 00:51:55.780 --> 00:51:58.600 align:middle line:84% I'm not saying it's optimal, but at least you want to open that. 00:51:58.600 --> 00:52:00.920 align:middle line:84% Just like we talked about, only in market failure 00:52:00.920 --> 00:52:02.540 align:middle line:90% should the government intervene. 00:52:02.540 --> 00:52:04.040 align:middle line:84% A Tiebout equilibrium is equilibrium 00:52:04.040 --> 00:52:05.060 align:middle line:90% with no market failures. 00:52:05.060 --> 00:52:06.100 align:middle line:84% So the government should intervene. 00:52:06.100 --> 00:52:07.460 align:middle line:84% Everybody's in the right town for them. 00:52:07.460 --> 00:52:09.002 align:middle line:84% The government should leave it alone. 00:52:09.002 --> 00:52:10.640 align:middle line:84% But if people are stuck and can't 00:52:10.640 --> 00:52:12.580 align:middle line:84% move to get to the Tiebout equilibrium-- 00:52:12.580 --> 00:52:15.127 align:middle line:84% so if, for example, someone has a family that really wants 00:52:15.127 --> 00:52:16.960 align:middle line:84% a good education for their kids, but they're 00:52:16.960 --> 00:52:18.840 align:middle line:84% stuck in a town with a shitty educational system, 00:52:18.840 --> 00:52:20.298 align:middle line:84% and they can't move to another town 00:52:20.298 --> 00:52:22.360 align:middle line:84% because zoning laws keep them out, 00:52:22.360 --> 00:52:23.580 align:middle line:90% then we should redistribute. 00:52:23.580 --> 00:52:24.080 align:middle line:90% Steven? 00:52:24.080 --> 00:52:25.580 align:middle line:84% AUDIENCE: I guess [INAUDIBLE], but I 00:52:25.580 --> 00:52:28.920 align:middle line:84% was going to ask like the difference between a poor town 00:52:28.920 --> 00:52:32.000 align:middle line:84% and a rich town in a Tiebout situation, 00:52:32.000 --> 00:52:35.000 align:middle line:84% could that [INAUDIBLE] be driven by someone in the poor town 00:52:35.000 --> 00:52:36.300 align:middle line:84% might actually want to live in the rich town, 00:52:36.300 --> 00:52:37.840 align:middle line:84% but they can't actually [INAUDIBLE]. 00:52:37.840 --> 00:52:38.120 align:middle line:90% JON GRUBER: Right. 00:52:38.120 --> 00:52:38.260 align:middle line:90% Right. 00:52:38.260 --> 00:52:41.180 align:middle line:84% So that would be a breakdown of the [? table. ?] So essentially, 00:52:41.180 --> 00:52:45.380 align:middle line:84% what's pretty cool about this is essentially if someone asks you, 00:52:45.380 --> 00:52:47.983 align:middle line:84% as people often come walk up and ask you, they'll say, hey, 00:52:47.983 --> 00:52:50.400 align:middle line:84% should we be redistributing from rich to poor communities? 00:52:50.400 --> 00:52:51.640 align:middle line:90% How do I think about that? 00:52:51.640 --> 00:52:53.460 align:middle line:84% And your answer is, it all depends 00:52:53.460 --> 00:52:56.180 align:middle line:84% on how efficiently we think how people are sorted 00:52:56.180 --> 00:52:57.277 align:middle line:90% across those communities. 00:52:57.277 --> 00:52:59.360 align:middle line:84% If we think we're in a purely Tiebout equilibrium, 00:52:59.360 --> 00:53:00.200 align:middle line:90% the answer is no. 00:53:00.200 --> 00:53:01.980 align:middle line:84% Given that we're not, the answer is probably a little bit 00:53:01.980 --> 00:53:03.320 align:middle line:90% yes and how much depends. 00:53:03.320 --> 00:53:06.210 align:middle line:90% 00:53:06.210 --> 00:53:09.500 align:middle line:84% So now let's say-- so let's say-- 00:53:09.500 --> 00:53:11.855 align:middle line:84% so let's say we've decided we do want to redistribute. 00:53:11.855 --> 00:53:13.480 align:middle line:84% So now the question is, how we do that? 00:53:13.480 --> 00:53:16.120 align:middle line:84% And the way we do it is through intergovernmental grants. 00:53:16.120 --> 00:53:17.620 align:middle line:84% So I want to talk about the analysis 00:53:17.620 --> 00:53:19.100 align:middle line:90% of intergovernmental grants. 00:53:19.100 --> 00:53:20.160 align:middle line:90% This is a fun example. 00:53:20.160 --> 00:53:23.580 align:middle line:84% It shows you how you have fun with budget constraints. 00:53:23.580 --> 00:53:30.500 align:middle line:84% So let's imagine that there's the town of Lexington, 00:53:30.500 --> 00:53:33.290 align:middle line:84% and we're in a simple world with a private good 00:53:33.290 --> 00:53:34.040 align:middle line:90% and a public good. 00:53:34.040 --> 00:53:35.790 align:middle line:84% There are private goods and a public good. 00:53:35.790 --> 00:53:37.180 align:middle line:90% The public good is education. 00:53:37.180 --> 00:53:39.180 align:middle line:84% So people can either spend the spend their money 00:53:39.180 --> 00:53:42.150 align:middle line:84% on private goods, or they can spend their money on schooling. 00:53:42.150 --> 00:53:44.655 align:middle line:84% And to make life easy, a unit of each is $1. 00:53:44.655 --> 00:53:46.030 align:middle line:84% We always do this numerator thing 00:53:46.030 --> 00:53:47.572 align:middle line:84% so we can talk about units of dollars 00:53:47.572 --> 00:53:51.590 align:middle line:84% interoperably without worrying about each. 00:53:51.590 --> 00:53:54.270 align:middle line:84% So let's assume people can-- the town of Lexington 00:53:54.270 --> 00:53:57.190 align:middle line:84% has a representative democracy, and it's 00:53:57.190 --> 00:54:00.330 align:middle line:84% choosing for its people-- and it's a true median voter theorem 00:54:00.330 --> 00:54:01.570 align:middle line:90% representative democracy. 00:54:01.570 --> 00:54:04.470 align:middle line:84% It's choosing for its people how much spend in education. 00:54:04.470 --> 00:54:06.950 align:middle line:84% So it's essentially representing its voters 00:54:06.950 --> 00:54:09.750 align:middle line:84% in thinking about that trade-off. 00:54:09.750 --> 00:54:12.670 align:middle line:84% Let's further assume that more education spending 00:54:12.670 --> 00:54:13.850 align:middle line:90% delivers a better outcome. 00:54:13.850 --> 00:54:14.810 align:middle line:90% Let's assume people want more. 00:54:14.810 --> 00:54:16.610 align:middle line:84% I'm going to come back to that next chapter. 00:54:16.610 --> 00:54:18.550 align:middle line:84% Let's just assume that every dollar of education spending 00:54:18.550 --> 00:54:19.950 align:middle line:90% is that much better outcome. 00:54:19.950 --> 00:54:21.570 align:middle line:84% So we want more education spending, 00:54:21.570 --> 00:54:23.670 align:middle line:90% but they also want other stuff. 00:54:23.670 --> 00:54:26.590 align:middle line:84% So what that means is that essentially you get something 00:54:26.590 --> 00:54:32.290 align:middle line:84% like figure 10-2 shows the trade-off facing Lexington. 00:54:32.290 --> 00:54:35.230 align:middle line:84% This is below table 10-2 that we looked at a few minutes ago-- 00:54:35.230 --> 00:54:38.103 align:middle line:84% shows the trade-off facing Lexington. 00:54:38.103 --> 00:54:39.770 align:middle line:84% People, let's say, can spend their money 00:54:39.770 --> 00:54:42.610 align:middle line:90% on private goods or education. 00:54:42.610 --> 00:54:50.930 align:middle line:84% Let's say the town of Lexington has $1,000 as an entity. 00:54:50.930 --> 00:54:54.370 align:middle line:84% And it can decide as a town whether they want to spend that 00:54:54.370 --> 00:54:57.190 align:middle line:84% $1,000 on private goods or education. 00:54:57.190 --> 00:54:59.690 align:middle line:84% And once again, we're ignoring the whole voting difficulties 00:54:59.690 --> 00:54:59.910 align:middle line:90% here. 00:54:59.910 --> 00:55:00.950 align:middle line:84% Let's assume it's a representative. 00:55:00.950 --> 00:55:02.575 align:middle line:84% It's representing people's preferences. 00:55:02.575 --> 00:55:05.090 align:middle line:84% You can assume everybody's identical to make this easy. 00:55:05.090 --> 00:55:09.630 align:middle line:84% So let's assume the town, as a set of identical individuals, 00:55:09.630 --> 00:55:12.650 align:middle line:84% has some indifference curve between private goods 00:55:12.650 --> 00:55:13.277 align:middle line:90% and education. 00:55:13.277 --> 00:55:15.610 align:middle line:84% And that indifference curve is tangent with their budget 00:55:15.610 --> 00:55:18.410 align:middle line:90% constraint at the point x. 00:55:18.410 --> 00:55:20.250 align:middle line:84% So let's say that the voters of Lexington 00:55:20.250 --> 00:55:22.710 align:middle line:84% decide they want to spend half their money on education, 00:55:22.710 --> 00:55:24.890 align:middle line:84% and half their money on everything else. 00:55:24.890 --> 00:55:27.610 align:middle line:90% And that's the equilibrium. 00:55:27.610 --> 00:55:31.210 align:middle line:84% Now, let's say the federal government says, you know what? 00:55:31.210 --> 00:55:34.725 align:middle line:84% We think you are not spending enough on education. 00:55:34.725 --> 00:55:36.150 align:middle line:90% [INAUDIBLE] 00:55:36.150 --> 00:55:37.910 align:middle line:90% We want you to spend more. 00:55:37.910 --> 00:55:40.030 align:middle line:84% And to entice you to spend more, we 00:55:40.030 --> 00:55:43.590 align:middle line:84% are going to give you a grant to try to entice you. 00:55:43.590 --> 00:55:46.430 align:middle line:84% There are three forms that grant can take. 00:55:46.430 --> 00:55:51.430 align:middle line:84% The first form is what we call a matching grant. 00:55:51.430 --> 00:55:53.950 align:middle line:84% A matching grant says, hey, Lexington, 00:55:53.950 --> 00:55:55.550 align:middle line:84% I'm going to pass a state law that 00:55:55.550 --> 00:55:57.970 align:middle line:84% says for every dollar you spend in education, 00:55:57.970 --> 00:55:59.910 align:middle line:90% I'm going to give you $1. 00:55:59.910 --> 00:56:02.910 align:middle line:84% So I'm going to match one-for-one your education 00:56:02.910 --> 00:56:04.438 align:middle line:90% spending. 00:56:04.438 --> 00:56:06.230 align:middle line:84% What does that do to the budget constraint? 00:56:06.230 --> 00:56:08.430 align:middle line:90% See that in figure 10-3. 00:56:08.430 --> 00:56:12.270 align:middle line:84% In figure 10-3, you see the new in red is the new budget 00:56:12.270 --> 00:56:14.430 align:middle line:90% constraint. 00:56:14.430 --> 00:56:17.150 align:middle line:84% This budget is such that if people in Lexington decide 00:56:17.150 --> 00:56:19.370 align:middle line:84% to spend all their money on private goods, 00:56:19.370 --> 00:56:22.910 align:middle line:84% they still only have $1,000 worth of private goods. 00:56:22.910 --> 00:56:26.770 align:middle line:84% But if they decide to spend all their money on public goods, 00:56:26.770 --> 00:56:29.170 align:middle line:84% they now get $2,000 worth of public goods. 00:56:29.170 --> 00:56:29.790 align:middle line:90% Why? 00:56:29.790 --> 00:56:34.010 align:middle line:84% Because the state has announced they're matching it one-for-one. 00:56:34.010 --> 00:56:35.787 align:middle line:84% So the budget curve has shifted outwards. 00:56:35.787 --> 00:56:37.370 align:middle line:84% In other words, the price of education 00:56:37.370 --> 00:56:40.990 align:middle line:84% has fallen to half the price of private goods. 00:56:40.990 --> 00:56:43.190 align:middle line:84% Education used to cost the same as private goods. 00:56:43.190 --> 00:56:46.997 align:middle line:84% Education now costs the town half as much. 00:56:46.997 --> 00:56:48.830 align:middle line:84% What will that do to the level of education? 00:56:48.830 --> 00:56:51.330 align:middle line:84% Well, we know from 14.01 that depends on income 00:56:51.330 --> 00:56:53.810 align:middle line:90% and substitution effects. 00:56:53.810 --> 00:56:57.250 align:middle line:84% On the one hand, the price has fallen. 00:56:57.250 --> 00:56:59.610 align:middle line:84% So substitution effects unambiguously 00:56:59.610 --> 00:57:02.530 align:middle line:84% lead you to say they want more of it. 00:57:02.530 --> 00:57:05.410 align:middle line:84% On the other hand, they're richer. 00:57:05.410 --> 00:57:09.610 align:middle line:84% So if education is a normal good, which we'll assume, 00:57:09.610 --> 00:57:14.450 align:middle line:84% that would say that they want more of it. 00:57:14.450 --> 00:57:17.170 align:middle line:84% The income effect would work the same way. 00:57:17.170 --> 00:57:19.490 align:middle line:84% They're now richer, so they have more of it as well. 00:57:19.490 --> 00:57:21.810 align:middle line:84% So the substitution effects would both say 00:57:21.810 --> 00:57:23.110 align:middle line:90% they want more education. 00:57:23.110 --> 00:57:25.730 align:middle line:84% Both the price has fallen, and they're richer. 00:57:25.730 --> 00:57:28.430 align:middle line:84% So for both reasons, they want more education. 00:57:28.430 --> 00:57:32.470 align:middle line:84% So education shifts out, and you end up 00:57:32.470 --> 00:57:35.710 align:middle line:84% spending $750 on education instead of $500. 00:57:35.710 --> 00:57:39.210 align:middle line:84% Note also, you also spend more money on private goods. 00:57:39.210 --> 00:57:39.710 align:middle line:90% Why? 00:57:39.710 --> 00:57:41.390 align:middle line:90% Because you're richer. 00:57:41.390 --> 00:57:43.370 align:middle line:84% And as long as private goods are normal, 00:57:43.370 --> 00:57:45.070 align:middle line:84% you're going to take that windfall 00:57:45.070 --> 00:57:47.670 align:middle line:90% and spread it across both. 00:57:47.670 --> 00:57:50.830 align:middle line:84% So you're going to spend $750 on education now, 00:57:50.830 --> 00:57:54.310 align:middle line:90% and $625 on private goods. 00:57:54.310 --> 00:57:56.230 align:middle line:84% So even though private goods price 00:57:56.230 --> 00:57:57.970 align:middle line:84% hasn't fallen since you're richer 00:57:57.970 --> 00:57:59.910 align:middle line:84% you want more private goods as well. 00:57:59.910 --> 00:58:04.550 align:middle line:84% So you are going to end up at a point like Y. Questions 00:58:04.550 --> 00:58:07.790 align:middle line:90% about that? 00:58:07.790 --> 00:58:10.430 align:middle line:84% Now, imagine the state said, well, wait a second. 00:58:10.430 --> 00:58:11.930 align:middle line:84% Let's try a different kind of grant. 00:58:11.930 --> 00:58:14.510 align:middle line:84% Let's give them what we call a block grant. 00:58:14.510 --> 00:58:22.550 align:middle line:84% A block grant is saying, look, we know that basically-- 00:58:22.550 --> 00:58:26.510 align:middle line:84% we know that that point you choose with the matching grant, 00:58:26.510 --> 00:58:34.360 align:middle line:84% if we gave you a block grant of $1,375, 00:58:34.360 --> 00:58:36.980 align:middle line:84% you could get to that exact point. 00:58:36.980 --> 00:58:37.480 align:middle line:90% Why? 00:58:37.480 --> 00:58:39.022 align:middle line:84% Because look at what you're spending. 00:58:39.022 --> 00:58:42.960 align:middle line:84% You're spending $750 plus $625 is $1,325. 00:58:42.960 --> 00:58:46.580 align:middle line:84% So if I literally just gave you $1,325, 00:58:46.580 --> 00:58:49.040 align:middle line:84% you would literally get to the same point. 00:58:49.040 --> 00:58:50.680 align:middle line:90% But is that what you do? 00:58:50.680 --> 00:58:51.800 align:middle line:90% It's not. 00:58:51.800 --> 00:58:54.160 align:middle line:90% Take a look at figure 10-4. 00:58:54.160 --> 00:58:58.000 align:middle line:84% Figure 10-4-- now in blue, we have the original budget 00:58:58.000 --> 00:59:02.880 align:middle line:84% constraint and the matching grant budget constraint. 00:59:02.880 --> 00:59:05.340 align:middle line:84% But we also have a third budget constraint, 00:59:05.340 --> 00:59:09.120 align:middle line:84% which is the block grant budget constraint now in red. 00:59:09.120 --> 00:59:15.600 align:middle line:84% That is just a shift out of the budget constraint of $1,375. 00:59:15.600 --> 00:59:20.640 align:middle line:84% Why does that cause a smaller increase in education spending 00:59:20.640 --> 00:59:23.040 align:middle line:90% than does the matching grant? 00:59:23.040 --> 00:59:25.140 align:middle line:84% Why does the block grant cause a smaller increase? 00:59:25.140 --> 00:59:25.940 align:middle line:90% Yeah? 00:59:25.940 --> 00:59:27.560 align:middle line:84% AUDIENCE: Would we only have the income effect? 00:59:27.560 --> 00:59:28.640 align:middle line:84% JON GRUBER: We only have the income effect. 00:59:28.640 --> 00:59:30.080 align:middle line:90% Relative prices haven't changed. 00:59:30.080 --> 00:59:31.660 align:middle line:84% This is like our old proof, remember, 00:59:31.660 --> 00:59:33.640 align:middle line:90% of income substitution effects. 00:59:33.640 --> 00:59:35.220 align:middle line:90% We only have the income effect. 00:59:35.220 --> 00:59:39.700 align:middle line:84% As education spending rises less, 00:59:39.700 --> 00:59:41.480 align:middle line:84% and private spending goes up more. 00:59:41.480 --> 00:59:45.100 align:middle line:90% 00:59:45.100 --> 00:59:49.505 align:middle line:84% Does that mean that the town is worse off? 00:59:49.505 --> 00:59:51.880 align:middle line:84% Is the town worse off or better off, and how do you know? 00:59:51.880 --> 00:59:57.100 align:middle line:90% 00:59:57.100 --> 00:59:58.600 align:middle line:84% Is the town worse off or better off, 00:59:58.600 --> 01:00:02.560 align:middle line:84% and how do you know under the block or of the matching grant? 01:00:02.560 --> 01:00:03.060 align:middle line:90% Yeah? 01:00:03.060 --> 01:00:04.380 align:middle line:90% AUDIENCE: [INAUDIBLE]. 01:00:04.380 --> 01:00:05.547 align:middle line:90% JON GRUBER: How do you know? 01:00:05.547 --> 01:00:06.492 align:middle line:90% AUDIENCE: [INAUDIBLE]. 01:00:06.492 --> 01:00:07.200 align:middle line:90% JON GRUBER: Yeah. 01:00:07.200 --> 01:00:09.060 align:middle line:84% By definition, it's better off because it could have chosen 01:00:09.060 --> 01:00:11.380 align:middle line:84% the exact same outcome as before, but it didn't. 01:00:11.380 --> 01:00:13.140 align:middle line:84% Or in other words, you can see the indifference curve 01:00:13.140 --> 01:00:15.765 align:middle line:84% tangent at Z will be higher than the indifference curve tangent 01:00:15.765 --> 01:00:18.820 align:middle line:84% at Y. That's the kind of reveal preference 01:00:18.820 --> 01:00:20.080 align:middle line:90% proof we like in economics. 01:00:20.080 --> 01:00:24.560 align:middle line:84% Just keep that in mind, is that if someone has a choice 01:00:24.560 --> 01:00:26.560 align:middle line:84% they made before and they choose not to make it, 01:00:26.560 --> 01:00:30.120 align:middle line:84% then they clearly like the new outcome better. 01:00:30.120 --> 01:00:34.880 align:middle line:84% So herein, we have the fundamental trade-off 01:00:34.880 --> 01:00:37.160 align:middle line:90% of grant-making. 01:00:37.160 --> 01:00:41.200 align:middle line:84% If your goal is just to increase education spending, 01:00:41.200 --> 01:00:45.240 align:middle line:84% then a better way to set that up is with a matching grant. 01:00:45.240 --> 01:00:48.360 align:middle line:84% If your goal is to make the town better off while also increasing 01:00:48.360 --> 01:00:51.640 align:middle line:84% education spending, then it's better to do the block grant. 01:00:51.640 --> 01:00:56.000 align:middle line:84% And that comes right back to why we're redistributing. 01:00:56.000 --> 01:00:59.675 align:middle line:84% If we feel like we really care about poor towns, 01:00:59.675 --> 01:01:01.300 align:middle line:84% and we also want education to increase, 01:01:01.300 --> 01:01:02.180 align:middle line:90% we may want a block grant. 01:01:02.180 --> 01:01:04.722 align:middle line:84% We may want to say, look, we realize education won't increase 01:01:04.722 --> 01:01:08.192 align:middle line:84% as much, but at least we'll give more to the poor towns. 01:01:08.192 --> 01:01:09.900 align:middle line:84% A matching grant gives more to rich towns 01:01:09.900 --> 01:01:12.320 align:middle line:84% because rich towns spend more on education. 01:01:12.320 --> 01:01:15.100 align:middle line:84% So a matching grant gives more to rich towns. 01:01:15.100 --> 01:01:17.480 align:middle line:84% A block grant gives the same to everyone. 01:01:17.480 --> 01:01:19.328 align:middle line:84% So if your goal is purely, we want 01:01:19.328 --> 01:01:20.620 align:middle line:90% to increase education spending. 01:01:20.620 --> 01:01:22.100 align:middle line:84% We don't care about redistribution, 01:01:22.100 --> 01:01:24.120 align:middle line:90% then a matching grant dominates. 01:01:24.120 --> 01:01:27.500 align:middle line:84% If your goal is both to increase educational spending 01:01:27.500 --> 01:01:31.980 align:middle line:84% and to redistribute across towns, a block grant's better. 01:01:31.980 --> 01:01:34.420 align:middle line:84% Now, you're probably thinking, well, Jon, 01:01:34.420 --> 01:01:36.260 align:middle line:84% can't we have our cake and eat it too? 01:01:36.260 --> 01:01:39.240 align:middle line:84% Can't we do a conditional block grant-- 01:01:39.240 --> 01:01:41.980 align:middle line:90% 01:01:41.980 --> 01:01:45.802 align:middle line:90% a conditional block grant? 01:01:45.802 --> 01:01:47.760 align:middle line:84% What would a conditional block grant look like? 01:01:47.760 --> 01:01:50.093 align:middle line:84% Well, it would say, look-- conditional block grant would 01:01:50.093 --> 01:01:55.020 align:middle line:84% say, look, we are going to say we're going to give you, 01:01:55.020 --> 01:01:56.382 align:middle line:90% Lexington three-- 01:01:56.382 --> 01:01:57.340 align:middle line:90% I shouldn't-- this is-- 01:01:57.340 --> 01:01:58.680 align:middle line:84% I didn't say-- don't give them that grant. 01:01:58.680 --> 01:01:59.360 align:middle line:90% They give them this grant. 01:01:59.360 --> 01:01:59.860 align:middle line:90% My bad. 01:01:59.860 --> 01:02:01.520 align:middle line:90% Wasn't $1,000. 01:02:01.520 --> 01:02:04.320 align:middle line:84% They give them $375 in that example [INAUDIBLE]. 01:02:04.320 --> 01:02:08.140 align:middle line:84% So in that example, they gave them $375, not $1,375. 01:02:08.140 --> 01:02:10.500 align:middle line:84% Because it increased their spending by $375 01:02:10.500 --> 01:02:13.020 align:middle line:90% is the delta they gave them. 01:02:13.020 --> 01:02:16.780 align:middle line:84% Well, now imagine that they gave them-- 01:02:16.780 --> 01:02:20.240 align:middle line:84% they said to Lexington, you can have $375, 01:02:20.240 --> 01:02:24.060 align:middle line:84% but you have to spend it on education. 01:02:24.060 --> 01:02:26.720 align:middle line:84% In that case, can we have our cake and eat it too? 01:02:26.720 --> 01:02:29.360 align:middle line:84% If we say to Lexington, $375, but you 01:02:29.360 --> 01:02:31.580 align:middle line:90% have to spend it on education. 01:02:31.580 --> 01:02:34.020 align:middle line:84% Will that both give a block grant, 01:02:34.020 --> 01:02:35.400 align:middle line:84% which is fair across communities, 01:02:35.400 --> 01:02:37.135 align:middle line:84% and enforce more education spending? 01:02:37.135 --> 01:02:39.760 align:middle line:84% And the answer is in economics, you never get to have your cake 01:02:39.760 --> 01:02:40.880 align:middle line:90% and eat it too. 01:02:40.880 --> 01:02:46.640 align:middle line:84% And to see, that let's go to figure 10-5, 01:02:46.640 --> 01:02:49.240 align:middle line:84% which should look familiar from our old figures about food 01:02:49.240 --> 01:02:51.800 align:middle line:90% stamps. 01:02:51.800 --> 01:02:55.653 align:middle line:84% Figure 10-5, we have the regular budget constraint B, 01:02:55.653 --> 01:02:56.820 align:middle line:90% the first budget constraint. 01:02:56.820 --> 01:03:00.240 align:middle line:84% We have the upper blue line is the matching grant. 01:03:00.240 --> 01:03:05.120 align:middle line:84% The line that goes DFE is the block grant. 01:03:05.120 --> 01:03:08.160 align:middle line:84% The new budget constraint goes AFE. 01:03:08.160 --> 01:03:09.660 align:middle line:90% Why? 01:03:09.660 --> 01:03:14.300 align:middle line:84% Why is the new budget constraint AFE, as opposed to the old DFE? 01:03:14.300 --> 01:03:16.840 align:middle line:90% 01:03:16.840 --> 01:03:19.540 align:middle line:84% Explain why you draw the new budget constraint that way. 01:03:19.540 --> 01:03:20.040 align:middle line:90% Yeah? 01:03:20.040 --> 01:03:21.352 align:middle line:90% AUDIENCE: [INAUDIBLE]. 01:03:21.352 --> 01:03:26.130 align:middle line:90% 01:03:26.130 --> 01:03:29.150 align:middle line:84% JON GRUBER: Yeah, the first $375 you have to spend on education. 01:03:29.150 --> 01:03:30.190 align:middle line:90% You don't have a choice. 01:03:30.190 --> 01:03:35.250 align:middle line:84% So that segment that goes from F to D is not available to you. 01:03:35.250 --> 01:03:37.310 align:middle line:84% You can't do-- that's not available to you. 01:03:37.310 --> 01:03:39.430 align:middle line:84% You have to spend at least $375 on education. 01:03:39.430 --> 01:03:41.950 align:middle line:84% Once you spend at least $375 on education, 01:03:41.950 --> 01:03:45.730 align:middle line:84% then you're back on the block grant. 01:03:45.730 --> 01:03:51.530 align:middle line:84% So what this means is, in many ways, 01:03:51.530 --> 01:03:55.710 align:middle line:84% you end up with the same outcome as a block grant, but not fully. 01:03:55.710 --> 01:03:56.568 align:middle line:90% And why is that? 01:03:56.568 --> 01:03:58.110 align:middle line:84% Why is it different than block grant? 01:03:58.110 --> 01:04:00.650 align:middle line:84% For what kind of towns will you not get the same outcome 01:04:00.650 --> 01:04:02.310 align:middle line:90% as the block grant? 01:04:02.310 --> 01:04:02.810 align:middle line:90% Yeah? 01:04:02.810 --> 01:04:04.735 align:middle line:84% AUDIENCE: Towns that have super high private goods spending 01:04:04.735 --> 01:04:05.235 align:middle line:90% [INAUDIBLE]. 01:04:05.235 --> 01:04:06.090 align:middle line:90% JON GRUBER: Yeah. 01:04:06.090 --> 01:04:09.330 align:middle line:84% You can see towns which might-- any town which would choose 01:04:09.330 --> 01:04:12.770 align:middle line:84% a point between D and F will now be forced 01:04:12.770 --> 01:04:18.010 align:middle line:84% to spend more on education and be on a lower indifference 01:04:18.010 --> 01:04:19.890 align:middle line:90% curve. 01:04:19.890 --> 01:04:22.807 align:middle line:84% If your tangency was a D and F and I forced-- between D and F, 01:04:22.807 --> 01:04:25.390 align:middle line:84% and I forced you to F, you'd be in a lower indifference curve. 01:04:25.390 --> 01:04:27.570 align:middle line:84% And that's probably the poorest towns. 01:04:27.570 --> 01:04:29.070 align:middle line:84% It's probably the poorest towns that 01:04:29.070 --> 01:04:30.590 align:middle line:90% spend the least on education. 01:04:30.590 --> 01:04:33.738 align:middle line:84% So in some sense, this is not as good for redistribution 01:04:33.738 --> 01:04:34.530 align:middle line:90% as the block grant. 01:04:34.530 --> 01:04:36.947 align:middle line:84% It's better for getting more education spending because it 01:04:36.947 --> 01:04:39.627 align:middle line:84% forces the towns that weren't spending on education to spend. 01:04:39.627 --> 01:04:41.210 align:middle line:84% But it's not as good for distribution. 01:04:41.210 --> 01:04:43.390 align:middle line:84% So it's sort of a halfway between example. 01:04:43.390 --> 01:04:45.910 align:middle line:84% So matching grant gets the maximum increase 01:04:45.910 --> 01:04:47.710 align:middle line:90% in education spending. 01:04:47.710 --> 01:04:51.353 align:middle line:84% Block grant gets the maximum increase in town utility. 01:04:51.353 --> 01:04:53.270 align:middle line:84% Conditional block grant is halfway in between, 01:04:53.270 --> 01:04:55.550 align:middle line:90% part way in between. 01:04:55.550 --> 01:04:57.950 align:middle line:90% Questions about that? 01:04:57.950 --> 01:04:58.630 align:middle line:90% Yeah? 01:04:58.630 --> 01:04:59.950 align:middle line:84% AUDIENCE: So like much like food stamps, 01:04:59.950 --> 01:05:01.350 align:middle line:84% poor towns would still be unhappy 01:05:01.350 --> 01:05:04.017 align:middle line:84% because their indifference curve gets lower, even though they're 01:05:04.017 --> 01:05:04.850 align:middle line:90% getting more money? 01:05:04.850 --> 01:05:05.790 align:middle line:90% JON GRUBER: Exactly. 01:05:05.790 --> 01:05:07.230 align:middle line:90% Exactly. 01:05:07.230 --> 01:05:07.950 align:middle line:90% Well, no. 01:05:07.950 --> 01:05:09.688 align:middle line:84% Lower-- they're not getting more money. 01:05:09.688 --> 01:05:11.730 align:middle line:84% The people between D and F are getting the same-- 01:05:11.730 --> 01:05:12.070 align:middle line:84% AUDIENCE: They're being forced to-- yeah. 01:05:12.070 --> 01:05:13.030 align:middle line:84% JON GRUBER: They're being forced to spend 01:05:13.030 --> 01:05:14.310 align:middle line:90% their money differently. 01:05:14.310 --> 01:05:16.310 align:middle line:90% Yeah, Steven? 01:05:16.310 --> 01:05:18.750 align:middle line:84% AUDIENCE: Does matching always strictly outperform 01:05:18.750 --> 01:05:22.850 align:middle line:84% the conditional block grant in terms of education [INAUDIBLE]? 01:05:22.850 --> 01:05:25.983 align:middle line:84% JON GRUBER: At the same budget level it will-- 01:05:25.983 --> 01:05:27.650 align:middle line:84% at the same budget level because you get 01:05:27.650 --> 01:05:29.210 align:middle line:84% the income and substitution effect, 01:05:29.210 --> 01:05:30.670 align:middle line:84% unless substitute effects are zero. 01:05:30.670 --> 01:05:32.140 align:middle line:84% So it's strictly as good or better 01:05:32.140 --> 01:05:33.890 align:middle line:84% in terms of education spending because you 01:05:33.890 --> 01:05:38.530 align:middle line:84% get substitution effect you don't get with any block grant. 01:05:38.530 --> 01:05:41.610 align:middle line:84% So basically, what you have is a situation-- 01:05:41.610 --> 01:05:44.730 align:middle line:90% let me just check that. 01:05:44.730 --> 01:05:46.165 align:middle line:90% At the same budget, yes. 01:05:46.165 --> 01:05:48.290 align:middle line:84% At the same budget, it will strictly outperform it. 01:05:48.290 --> 01:05:49.530 align:middle line:84% Now, matching grant at less money 01:05:49.530 --> 01:05:51.697 align:middle line:84% won't necessarily outperform it, but the same budget 01:05:51.697 --> 01:05:53.690 align:middle line:90% will strictly outperform it. 01:05:53.690 --> 01:05:57.210 align:middle line:84% So basically, what you have is a situation where 01:05:57.210 --> 01:06:00.350 align:middle line:84% there's this inherent trade-off, and it all comes full circle. 01:06:00.350 --> 01:06:03.257 align:middle line:84% What I like about this chapter, it's very circular, 01:06:03.257 --> 01:06:05.090 align:middle line:84% in the sense it all comes full circle, which 01:06:05.090 --> 01:06:09.690 align:middle line:84% is it all comes back to Tiebout, which is if Tiebout is holding 01:06:09.690 --> 01:06:13.730 align:middle line:84% perfectly, we just say, isn't it wonderful local public goods are 01:06:13.730 --> 01:06:14.850 align:middle line:90% efficient? 01:06:14.850 --> 01:06:16.290 align:middle line:90% That's wonderful. 01:06:16.290 --> 01:06:19.950 align:middle line:84% But whether that's going to be true depends on a lot of things. 01:06:19.950 --> 01:06:21.990 align:middle line:84% What you should do at which level 01:06:21.990 --> 01:06:24.870 align:middle line:84% is going to depend on how much Tiebout holds. 01:06:24.870 --> 01:06:27.630 align:middle line:84% And whether you want to redistribute across communities 01:06:27.630 --> 01:06:29.070 align:middle line:84% and which kind of grant you use is 01:06:29.070 --> 01:06:31.350 align:middle line:84% going to depend on how much Tiebout holds. 01:06:31.350 --> 01:06:35.630 align:middle line:84% So it's a nice way to think about optimal fiscal federalism. 01:06:35.630 --> 01:06:39.558 align:middle line:90% Now, questions about that? 01:06:39.558 --> 01:06:41.850 align:middle line:84% We're going to do-- next time-- we're not stopping yet. 01:06:41.850 --> 01:06:43.475 align:middle line:84% But next time, we're going to come back 01:06:43.475 --> 01:06:47.230 align:middle line:84% to one last cool coda to all this, 01:06:47.230 --> 01:06:51.310 align:middle line:84% which is actually-- actually, no, I'm going to do it now. 01:06:51.310 --> 01:06:54.910 align:middle line:84% So basically, let's talk about two more things. 01:06:54.910 --> 01:06:57.630 align:middle line:84% The first other thing I want to talk about in chapter 10 01:06:57.630 --> 01:07:00.270 align:middle line:84% is an important public policy called school finance 01:07:00.270 --> 01:07:02.430 align:middle line:90% equalization. 01:07:02.430 --> 01:07:07.270 align:middle line:90% Many states have felt that-- 01:07:07.270 --> 01:07:10.270 align:middle line:84% first of all, there's enormous discrepancy across districts 01:07:10.270 --> 01:07:12.070 align:middle line:84% and what they spend in education. 01:07:12.070 --> 01:07:15.750 align:middle line:84% So within Texas, for example, there's 01:07:15.750 --> 01:07:18.130 align:middle line:84% a five-fold discrepancy across towns 01:07:18.130 --> 01:07:21.510 align:middle line:84% and what they spend on education per student. 01:07:21.510 --> 01:07:22.870 align:middle line:90% That's true all over the place. 01:07:22.870 --> 01:07:25.530 align:middle line:90% Huge discrepancy. 01:07:25.530 --> 01:07:31.730 align:middle line:84% So What happened was people started 01:07:31.730 --> 01:07:35.290 align:middle line:84% suing under state constitutions, saying 01:07:35.290 --> 01:07:37.650 align:middle line:84% that the state Constitution guaranteed equal rights 01:07:37.650 --> 01:07:40.690 align:middle line:84% for everyone, and different-- such vast disparities 01:07:40.690 --> 01:07:43.570 align:middle line:84% in educational spending violated their equal rights. 01:07:43.570 --> 01:07:48.210 align:middle line:84% And the first such case was won, once again, 01:07:48.210 --> 01:07:51.130 align:middle line:90% in California in 1976. 01:07:51.130 --> 01:07:54.490 align:middle line:84% In California, in 1976, there was a lawsuit. 01:07:54.490 --> 01:07:58.370 align:middle line:84% And the state ruled that the state-- the Supreme Court 01:07:58.370 --> 01:08:01.410 align:middle line:84% of the state ruled that the state had to equalize financing 01:08:01.410 --> 01:08:03.337 align:middle line:90% across districts. 01:08:03.337 --> 01:08:05.170 align:middle line:84% And indeed, there's been subsequent lawsuits 01:08:05.170 --> 01:08:07.530 align:middle line:84% in almost every state in the country, 01:08:07.530 --> 01:08:10.570 align:middle line:84% and that's led to a massive shift towards school finance 01:08:10.570 --> 01:08:11.213 align:middle line:90% equalization. 01:08:11.213 --> 01:08:12.630 align:middle line:84% I haven't gotten to education yet, 01:08:12.630 --> 01:08:15.250 align:middle line:84% but this is how education is paid for. 01:08:15.250 --> 01:08:18.069 align:middle line:90% Now, these plans differ a lot. 01:08:18.069 --> 01:08:22.430 align:middle line:84% For example, in California, what California says is, look, 01:08:22.430 --> 01:08:25.750 align:middle line:84% every district should spend the same. 01:08:25.750 --> 01:08:29.910 align:middle line:84% You can exceed that by $350 per student, which is tiny. 01:08:29.910 --> 01:08:32.910 align:middle line:84% But beyond that, you don't get to have any more. 01:08:32.910 --> 01:08:34.950 align:middle line:84% So in California, basically, it doesn't 01:08:34.950 --> 01:08:37.149 align:middle line:84% matter how much you collect in property taxes, 01:08:37.149 --> 01:08:40.029 align:middle line:84% your education spending can only be $350 01:08:40.029 --> 01:08:43.510 align:middle line:90% above the statewide average. 01:08:43.510 --> 01:08:46.310 align:middle line:84% New Jersey, on the other hand, does a different system where 01:08:46.310 --> 01:08:53.189 align:middle line:84% they say, look, if you are below the 85th percentile of property 01:08:53.189 --> 01:08:55.710 align:middle line:84% values in the state, we're going to send you money. 01:08:55.710 --> 01:08:57.503 align:middle line:90% And above, we're not. 01:08:57.503 --> 01:08:59.670 align:middle line:84% So a different system-- both redistribution but done 01:08:59.670 --> 01:09:01.710 align:middle line:90% in very different ways. 01:09:01.710 --> 01:09:03.770 align:middle line:84% What are the effects of this redistribution? 01:09:03.770 --> 01:09:05.189 align:middle line:90% It's quite interesting. 01:09:05.189 --> 01:09:06.649 align:middle line:84% There's been a number of studies. 01:09:06.649 --> 01:09:07.691 align:middle line:90% They found the following. 01:09:07.691 --> 01:09:10.990 align:middle line:84% First of all, all these schemes equalize school financing. 01:09:10.990 --> 01:09:12.830 align:middle line:90% All these schemes worked. 01:09:12.830 --> 01:09:15.359 align:middle line:84% The spending across districts got much more equal, 01:09:15.359 --> 01:09:17.439 align:middle line:84% but how they worked was very different. 01:09:17.439 --> 01:09:20.920 align:middle line:84% States like New Jersey, they work by bringing the bottom up. 01:09:20.920 --> 01:09:25.200 align:middle line:84% States like California, they work by bringing the top down. 01:09:25.200 --> 01:09:25.840 align:middle line:90% Why? 01:09:25.840 --> 01:09:27.840 align:middle line:84% Because think about the California system. 01:09:27.840 --> 01:09:30.200 align:middle line:90% It's essentially 100% tax. 01:09:30.200 --> 01:09:33.200 align:middle line:84% Once you're spending $350 above the average, 01:09:33.200 --> 01:09:35.240 align:middle line:84% all the rest of the tax revenues you raise just 01:09:35.240 --> 01:09:38.410 align:middle line:84% go to pay for other towns' education. 01:09:38.410 --> 01:09:39.660 align:middle line:90% So people are like, forget it. 01:09:39.660 --> 01:09:41.319 align:middle line:84% We're not going to invest in education. 01:09:41.319 --> 01:09:44.899 align:middle line:84% Whereas, in New Jersey, the rich towns get taxed, 01:09:44.899 --> 01:09:45.800 align:middle line:90% but they still-- 01:09:45.800 --> 01:09:47.680 align:middle line:90% it's like a typical tax. 01:09:47.680 --> 01:09:50.180 align:middle line:84% Some of what they spend they have to give to the government, 01:09:50.180 --> 01:09:51.300 align:middle line:90% but the rest they get to keep. 01:09:51.300 --> 01:09:52.479 align:middle line:84% So they might spend less on education 01:09:52.479 --> 01:09:53.562 align:middle line:90% than they otherwise would. 01:09:53.562 --> 01:09:57.560 align:middle line:84% But they redistribute it to the bottom, which spends a lot more. 01:09:57.560 --> 01:10:00.640 align:middle line:84% So there's equalization, but the way it happens 01:10:00.640 --> 01:10:03.320 align:middle line:84% is very different depending how the system is set up. 01:10:03.320 --> 01:10:06.640 align:middle line:84% So this is-- it's not quite pedantic. 01:10:06.640 --> 01:10:10.520 align:middle line:84% It's a technical detail, but it matters enormously. 01:10:10.520 --> 01:10:13.660 align:middle line:84% And this matters more than that because studies 01:10:13.660 --> 01:10:15.760 align:middle line:84% have shown this matters for educational outcomes. 01:10:15.760 --> 01:10:18.540 align:middle line:84% And we'll talk about money and education next time. 01:10:18.540 --> 01:10:21.457 align:middle line:84% But studies have shown which not only does equalization lead 01:10:21.457 --> 01:10:23.540 align:middle line:84% to equalization of spending, at least equalization 01:10:23.540 --> 01:10:25.380 align:middle line:90% of student outcomes. 01:10:25.380 --> 01:10:26.940 align:middle line:90% This stuff matters. 01:10:26.940 --> 01:10:32.780 align:middle line:84% So that's an example of how equalization works. 01:10:32.780 --> 01:10:35.420 align:middle line:84% Now, this leads to one of my favorite examples in this 01:10:35.420 --> 01:10:38.800 align:middle line:84% course, which is someone wrote an article-- 01:10:38.800 --> 01:10:43.140 align:middle line:84% this guy wrote this article in the 1980s about why did 01:10:43.140 --> 01:10:46.140 align:middle line:90% Proposition 13 happen? 01:10:46.140 --> 01:10:48.420 align:middle line:84% Proposition 13 was this incredible event 01:10:48.420 --> 01:10:51.260 align:middle line:84% that changed the whole nature of public finance in the country. 01:10:51.260 --> 01:10:53.580 align:middle line:84% But people have bitched about property taxes 01:10:53.580 --> 01:10:55.780 align:middle line:90% since the first century. 01:10:55.780 --> 01:10:58.960 align:middle line:90% Why 1978? 01:10:58.960 --> 01:11:02.180 align:middle line:84% Why did Proposition 13 not happen until 1978? 01:11:02.180 --> 01:11:03.900 align:middle line:84% And he said, well, the answer is, what 01:11:03.900 --> 01:11:06.380 align:middle line:90% happened in California in 1976? 01:11:06.380 --> 01:11:08.500 align:middle line:90% School finance equalization? 01:11:08.500 --> 01:11:10.620 align:middle line:84% What does school finance equalization do? 01:11:10.620 --> 01:11:14.240 align:middle line:84% It fundamentally breaks the tax benefit linkage 01:11:14.240 --> 01:11:16.880 align:middle line:84% between the taxes you collect and what you get from them. 01:11:16.880 --> 01:11:21.000 align:middle line:84% Before 1976, if you raised more tax, you got better schools. 01:11:21.000 --> 01:11:22.540 align:middle line:84% After 1976, you raised more taxes, 01:11:22.540 --> 01:11:23.748 align:middle line:90% you don't get better schools. 01:11:23.748 --> 01:11:25.520 align:middle line:90% So why pay higher taxes? 01:11:25.520 --> 01:11:28.900 align:middle line:84% So his argument-- this is more of a curiosity argument. 01:11:28.900 --> 01:11:31.640 align:middle line:90% It's not really strongly proven. 01:11:31.640 --> 01:11:35.040 align:middle line:84% His argument is that basically, Proposition 13 01:11:35.040 --> 01:11:39.920 align:middle line:84% was a natural consequence of causing-- 01:11:39.920 --> 01:11:43.920 align:middle line:84% Essentially, what the school finance equalization in 1976 was 01:11:43.920 --> 01:11:47.540 align:middle line:84% impose 100% tax on your property taxes, 01:11:47.540 --> 01:11:50.875 align:middle line:84% giving it back to the state, breaking the tax benefit linkage 01:11:50.875 --> 01:11:52.500 align:middle line:84% and causing towns say, well, forget it. 01:11:52.500 --> 01:11:55.400 align:middle line:84% We don't want property taxes anymore. 01:11:55.400 --> 01:12:00.360 align:middle line:84% That led to all the property taxes falling, 01:12:00.360 --> 01:12:03.840 align:middle line:84% which led to the entire educational system cratering. 01:12:03.840 --> 01:12:05.220 align:middle line:90% Somebody mentioned a cycle. 01:12:05.220 --> 01:12:07.400 align:middle line:90% This is a real cycle, you see. 01:12:07.400 --> 01:12:09.780 align:middle line:90% You break the linkage. 01:12:09.780 --> 01:12:12.540 align:middle line:84% People vote for lower taxes, and the whole bottom 01:12:12.540 --> 01:12:14.000 align:middle line:90% falls out of the school system. 01:12:14.000 --> 01:12:15.700 align:middle line:84% And that's happened in California. 01:12:15.700 --> 01:12:19.740 align:middle line:84% Really fascinating set of events in the late 1970s. 01:12:19.740 --> 01:12:20.580 align:middle line:90% Yeah? 01:12:20.580 --> 01:12:23.020 align:middle line:84% AUDIENCE: Can I ask if the law in California 01:12:23.020 --> 01:12:26.500 align:middle line:84% is causing overall spending to drop, why is it still 01:12:26.500 --> 01:12:27.940 align:middle line:90% the same way [INAUDIBLE]? 01:12:27.940 --> 01:12:29.320 align:middle line:90% JON GRUBER: It is very hard. 01:12:29.320 --> 01:12:31.620 align:middle line:84% It turns out to be very-- it's a great question. 01:12:31.620 --> 01:12:36.060 align:middle line:84% It turns to be very hard to make the argument 01:12:36.060 --> 01:12:38.760 align:middle line:84% that you should raise taxes, fundamentally. 01:12:38.760 --> 01:12:39.640 align:middle line:90% It's just very hard. 01:12:39.640 --> 01:12:42.300 align:middle line:84% I mean, you can go in and say, look, look 01:12:42.300 --> 01:12:44.100 align:middle line:90% what's happened to our schools. 01:12:44.100 --> 01:12:46.238 align:middle line:84% And the problem is, while you say that, people say, 01:12:46.238 --> 01:12:47.780 align:middle line:84% well, raising taxes wouldn't solve it 01:12:47.780 --> 01:12:49.822 align:middle line:84% with this corrupt politicians, da, da, da, da da. 01:12:49.822 --> 01:12:52.320 align:middle line:84% It's hard to change the status quo is the bottom line. 01:12:52.320 --> 01:12:56.300 align:middle line:84% So Massachusetts has a law like this called Proposition 2 1/2, 01:12:56.300 --> 01:13:01.620 align:middle line:84% which limits property taxes to growing at 2.5% a year. 01:13:01.620 --> 01:13:03.300 align:middle line:84% By the way, these limits are binding. 01:13:03.300 --> 01:13:05.360 align:middle line:84% In California-- here's a fascinating case. 01:13:05.360 --> 01:13:08.840 align:middle line:84% California, starting 1978, your property tax bill 01:13:08.840 --> 01:13:10.680 align:middle line:90% could only grow at 2% a year. 01:13:10.680 --> 01:13:14.868 align:middle line:84% California property values have gone up 50-fold since 1978. 01:13:14.868 --> 01:13:17.160 align:middle line:84% So you have houses that are worth gazillions of dollars 01:13:17.160 --> 01:13:19.020 align:middle line:84% paying almost nothing in property tax. 01:13:19.020 --> 01:13:22.092 align:middle line:84% Here's the fascinating thing-- until you sell the house. 01:13:22.092 --> 01:13:23.800 align:middle line:84% When you sell the house, the property tax 01:13:23.800 --> 01:13:25.983 align:middle line:84% gets reset at 1% of the new value. 01:13:25.983 --> 01:13:27.400 align:middle line:84% So that leads to this thing called 01:13:27.400 --> 01:13:29.000 align:middle line:84% housing lock, where you have people who don't 01:13:29.000 --> 01:13:30.240 align:middle line:90% want to sell their houses. 01:13:30.240 --> 01:13:31.740 align:middle line:84% Because basically, once they do it, 01:13:31.740 --> 01:13:33.863 align:middle line:84% there's a huge property tax hit the next owner. 01:13:33.863 --> 01:13:34.780 align:middle line:90% It's like a good deal. 01:13:34.780 --> 01:13:38.500 align:middle line:84% It's like, remember we talked about the old pollution, 01:13:38.500 --> 01:13:39.890 align:middle line:90% the old plants. 01:13:39.890 --> 01:13:41.640 align:middle line:84% And we had these new standards which said, 01:13:41.640 --> 01:13:43.720 align:middle line:84% only new plants have to put in scrubbers. 01:13:43.720 --> 01:13:46.425 align:middle line:84% Well, the new thing which says only new houses pay a higher 01:13:46.425 --> 01:13:48.300 align:middle line:84% property tax, people don't sell their houses. 01:13:48.300 --> 01:13:50.880 align:middle line:84% Another example of unanticipated distortion that 01:13:50.880 --> 01:13:52.533 align:middle line:90% comes from systems like this. 01:13:52.533 --> 01:13:53.700 align:middle line:90% But it turns out to be hard. 01:13:53.700 --> 01:13:56.400 align:middle line:84% So in Massachusetts, we have Proposition 2 1/2. 01:13:56.400 --> 01:13:58.400 align:middle line:84% But in the law, there's the ability 01:13:58.400 --> 01:14:00.880 align:middle line:90% of towns to have override votes. 01:14:00.880 --> 01:14:03.208 align:middle line:84% So towns can have specific votes to say, 01:14:03.208 --> 01:14:05.000 align:middle line:84% this year we're going to raise our property 01:14:05.000 --> 01:14:08.410 align:middle line:90% taxes extra to fund x. 01:14:08.410 --> 01:14:11.370 align:middle line:90% Why does that work? 01:14:11.370 --> 01:14:14.210 align:middle line:90% Those pass more than they fail. 01:14:14.210 --> 01:14:17.210 align:middle line:84% So even though Massachusetts kept Proposition 2 1/2, 01:14:17.210 --> 01:14:20.210 align:middle line:84% because they make it at every vote about a specific tax 01:14:20.210 --> 01:14:21.290 align:middle line:90% benefit linkage. 01:14:21.290 --> 01:14:23.370 align:middle line:84% So if California, I would argue, could 01:14:23.370 --> 01:14:25.940 align:middle line:84% prove that we promise every dollar of higher tax 01:14:25.940 --> 01:14:27.690 align:middle line:84% will go to better education, they probably 01:14:27.690 --> 01:14:29.123 align:middle line:90% could change their system. 01:14:29.123 --> 01:14:30.290 align:middle line:90% But it's hard to prove that. 01:14:30.290 --> 01:14:33.650 align:middle line:84% In Lexington, what we do is we have relatively modest increases 01:14:33.650 --> 01:14:36.070 align:middle line:84% for specific projects, like to fix this high school, 01:14:36.070 --> 01:14:37.490 align:middle line:90% to fix this building. 01:14:37.490 --> 01:14:39.830 align:middle line:84% So you maximize tax benefit linkages, 01:14:39.830 --> 01:14:41.710 align:middle line:84% and that's how you get around the law. 01:14:41.710 --> 01:14:42.710 align:middle line:90% It's hard to get around. 01:14:42.710 --> 01:14:44.010 align:middle line:84% But just vaguely, we should increase tax 01:14:44.010 --> 01:14:45.270 align:middle line:90% to make education better. 01:14:45.270 --> 01:14:48.150 align:middle line:84% It's hard to prove and hard for people to accept. 01:14:48.150 --> 01:14:49.010 align:middle line:90% AUDIENCE: Thank you. 01:14:49.010 --> 01:14:49.530 align:middle line:90% JON GRUBER: All right. 01:14:49.530 --> 01:14:50.488 align:middle line:90% That's a good question. 01:14:50.488 --> 01:14:53.130 align:middle line:90% Other questions? 01:14:53.130 --> 01:14:55.130 align:middle line:84% I like chapter 10 a lot because, once again, 01:14:55.130 --> 01:14:57.430 align:middle line:84% like chapter 8, on cost-benefit analysis, 01:14:57.430 --> 01:14:59.430 align:middle line:84% seems like it's getting kind of boring going in. 01:14:59.430 --> 01:15:01.310 align:middle line:84% It actually ends up being pretty interesting. 01:15:01.310 --> 01:15:02.852 align:middle line:84% Chapter 11 has a lot in it, so I want 01:15:02.852 --> 01:15:05.130 align:middle line:84% to get started on chapter 11 for five minutes. 01:15:05.130 --> 01:15:10.243 align:middle line:84% So let's take your chapter 11 handout, which 01:15:10.243 --> 01:15:11.910 align:middle line:84% is the other handout you have, and let's 01:15:11.910 --> 01:15:14.630 align:middle line:84% talk a little bit about education in the US. 01:15:14.630 --> 01:15:17.750 align:middle line:84% Education is the single largest expenditure for state and local 01:15:17.750 --> 01:15:18.410 align:middle line:90% governments. 01:15:18.410 --> 01:15:21.030 align:middle line:84% One third of all of our state and local budgets 01:15:21.030 --> 01:15:22.510 align:middle line:90% go to education. 01:15:22.510 --> 01:15:25.270 align:middle line:84% Yet, our education system is not very good. 01:15:25.270 --> 01:15:31.790 align:middle line:84% If you look at figure 11-2, this shows education spending 01:15:31.790 --> 01:15:35.110 align:middle line:84% and test scores on an internationally comparable 01:15:35.110 --> 01:15:36.990 align:middle line:90% eighth grade math test. 01:15:36.990 --> 01:15:40.790 align:middle line:84% What you'll notice is, first of all, the US spends the most 01:15:40.790 --> 01:15:43.905 align:middle line:90% and does the worst. 01:15:43.905 --> 01:15:46.030 align:middle line:84% Second of all, there's a remarkably low correlation 01:15:46.030 --> 01:15:48.337 align:middle line:84% between the blue bars and the orange bars-- 01:15:48.337 --> 01:15:50.670 align:middle line:84% a remarkably low correlation between who spends the most 01:15:50.670 --> 01:15:53.250 align:middle line:90% and who does the best. 01:15:53.250 --> 01:15:55.732 align:middle line:84% So this is-- I said we'd assume in the example 01:15:55.732 --> 01:15:58.190 align:middle line:84% a few minutes ago that more education spending meant better 01:15:58.190 --> 01:15:58.850 align:middle line:90% outcomes. 01:15:58.850 --> 01:16:00.560 align:middle line:90% This one doesn't look like it. 01:16:00.560 --> 01:16:02.810 align:middle line:84% It certainly doesn't look like more education spending 01:16:02.810 --> 01:16:03.850 align:middle line:90% is delivering better outcomes. 01:16:03.850 --> 01:16:05.730 align:middle line:84% There's no correlation between the blue bars 01:16:05.730 --> 01:16:06.850 align:middle line:90% and the orange bars. 01:16:06.850 --> 01:16:11.250 align:middle line:84% And that's why we raised this question of, 01:16:11.250 --> 01:16:14.170 align:middle line:84% is the government getting it wrong with education, and so 01:16:14.170 --> 01:16:15.930 align:middle line:90% education spending. 01:16:15.930 --> 01:16:18.610 align:middle line:84% So what I want to start with the last five minutes, 01:16:18.610 --> 01:16:20.160 align:middle line:90% I want to ask-- 01:16:20.160 --> 01:16:21.910 align:middle line:84% let's step back and ask the first question 01:16:21.910 --> 01:16:23.370 align:middle line:90% in public finance. 01:16:23.370 --> 01:16:25.452 align:middle line:84% Why is the government even in education? 01:16:25.452 --> 01:16:27.410 align:middle line:84% Why don't we just have a system where everybody 01:16:27.410 --> 01:16:29.330 align:middle line:90% pays for their own education? 01:16:29.330 --> 01:16:31.610 align:middle line:84% You could take out a loan from the bank, 01:16:31.610 --> 01:16:33.170 align:middle line:90% and pay for your own education. 01:16:33.170 --> 01:16:34.470 align:middle line:84% Why don't we just have private schools, 01:16:34.470 --> 01:16:35.830 align:middle line:84% everybody paying for their own education? 01:16:35.830 --> 01:16:36.570 align:middle line:90% Why not? 01:16:36.570 --> 01:16:39.090 align:middle line:84% What are the market failures in that system? 01:16:39.090 --> 01:16:40.530 align:middle line:84% So the private market system would 01:16:40.530 --> 01:16:44.450 align:middle line:84% be literally-- there's no public involvement in education. 01:16:44.450 --> 01:16:48.090 align:middle line:84% Literally, we just have everyone financing 01:16:48.090 --> 01:16:49.430 align:middle line:90% their own private education. 01:16:49.430 --> 01:16:51.210 align:middle line:90% What would be wrong with that? 01:16:51.210 --> 01:16:51.770 align:middle line:90% Nick? 01:16:51.770 --> 01:16:52.810 align:middle line:90% AUDIENCE: Externalities. 01:16:52.810 --> 01:16:55.850 align:middle line:84% JON GRUBER: Externalities of what type? 01:16:55.850 --> 01:16:58.690 align:middle line:84% AUDIENCE: Probably both, but positive externalities 01:16:58.690 --> 01:17:00.370 align:middle line:90% in the sense that-- 01:17:00.370 --> 01:17:03.248 align:middle line:84% yeah, positive externalities because it benefits everyone 01:17:03.248 --> 01:17:04.790 align:middle line:84% to have an [? educated population. ?] 01:17:04.790 --> 01:17:05.230 align:middle line:90% JON GRUBER: Yeah. 01:17:05.230 --> 01:17:07.272 align:middle line:84% So basically, there's two types of externalities. 01:17:07.272 --> 01:17:10.190 align:middle line:84% One is fiscal externalities, which is you pay higher taxes, 01:17:10.190 --> 01:17:12.690 align:middle line:84% and that benefits all of us if you have better education. 01:17:12.690 --> 01:17:16.215 align:middle line:84% But more generally, we think that citizenship, 01:17:16.215 --> 01:17:17.590 align:middle line:84% the basic tenets of what it means 01:17:17.590 --> 01:17:19.257 align:middle line:84% to be a citizen in the US, what it means 01:17:19.257 --> 01:17:21.670 align:middle line:84% to be a resident in the US are things we fundamentally 01:17:21.670 --> 01:17:22.610 align:middle line:90% want people to know. 01:17:22.610 --> 01:17:25.070 align:middle line:90% Remember, your vote affects me. 01:17:25.070 --> 01:17:29.310 align:middle line:84% I want you to know enough to cast a vote that's informed. 01:17:29.310 --> 01:17:34.110 align:middle line:84% There's a basic set of things that we want citizens to know. 01:17:34.110 --> 01:17:37.730 align:middle line:90% Now, so that's one benefit. 01:17:37.730 --> 01:17:43.390 align:middle line:84% Another externality benefit is things like reduced crime. 01:17:43.390 --> 01:17:46.590 align:middle line:84% We don't want people to be committing crimes. 01:17:46.590 --> 01:17:48.190 align:middle line:90% That affects all of us. 01:17:48.190 --> 01:17:49.990 align:middle line:84% The problem with those arguments is 01:17:49.990 --> 01:17:52.690 align:middle line:84% those are pretty compelling arguments for elementary school, 01:17:52.690 --> 01:17:54.630 align:middle line:84% middle school, maybe even high school. 01:17:54.630 --> 01:17:58.370 align:middle line:84% It's a little hard to justify AP classes. 01:17:58.370 --> 01:18:02.290 align:middle line:84% Like, I'd be a criminal if I hadn't taken AP chemistry. 01:18:02.290 --> 01:18:05.210 align:middle line:84% It's a little hard to justify a lot of what we spend money 01:18:05.210 --> 01:18:07.890 align:middle line:90% on education with those alone. 01:18:07.890 --> 01:18:09.970 align:middle line:90% So what else? 01:18:09.970 --> 01:18:11.250 align:middle line:90% Yeah? 01:18:11.250 --> 01:18:12.970 align:middle line:84% AUDIENCE: Jobs, entrepreneurship. 01:18:12.970 --> 01:18:15.650 align:middle line:84% JON GRUBER: Productivity writ large. 01:18:15.650 --> 01:18:20.330 align:middle line:84% But under what condition should we care? 01:18:20.330 --> 01:18:22.170 align:middle line:84% Under what condition should that cause 01:18:22.170 --> 01:18:24.410 align:middle line:84% us to want to publicly finance education? 01:18:24.410 --> 01:18:26.420 align:middle line:84% Or in which case, should it not want us to? 01:18:26.420 --> 01:18:28.670 align:middle line:84% Are you going to answer that, or a different question, 01:18:28.670 --> 01:18:29.420 align:middle line:90% different comment? 01:18:29.420 --> 01:18:30.995 align:middle line:90% AUDIENCE: [INAUDIBLE]. 01:18:30.995 --> 01:18:31.870 align:middle line:90% JON GRUBER: Well, no. 01:18:31.870 --> 01:18:33.578 align:middle line:84% I want to answer this question right now. 01:18:33.578 --> 01:18:35.650 align:middle line:90% Why might she be wrong? 01:18:35.650 --> 01:18:36.930 align:middle line:90% Why might the fact-- yeah? 01:18:36.930 --> 01:18:38.210 align:middle line:84% AUDIENCE: If I can fully capture my-- 01:18:38.210 --> 01:18:38.930 align:middle line:90% JON GRUBER: Yeah. 01:18:38.930 --> 01:18:40.890 align:middle line:84% Remember, it has to be an externality. 01:18:40.890 --> 01:18:43.170 align:middle line:84% Now, obviously you can't because you pay taxes. 01:18:43.170 --> 01:18:46.330 align:middle line:84% But beyond that, the question is, if I literally-- 01:18:46.330 --> 01:18:49.130 align:middle line:84% does Elon Musk generate more utility for the world 01:18:49.130 --> 01:18:51.290 align:middle line:84% than he captures himself in his private wealth, 01:18:51.290 --> 01:18:53.690 align:middle line:84% despite it-- maybe now not so much anymore. 01:18:53.690 --> 01:18:56.770 align:middle line:84% But five years ago, I would say yes. 01:18:56.770 --> 01:18:58.130 align:middle line:90% Elon Musk is worth $100 billion. 01:18:58.130 --> 01:19:00.380 align:middle line:84% The world is more than $100 billion better off because 01:19:00.380 --> 01:19:03.140 align:middle line:90% of Tesla and SpaceX and stuff. 01:19:03.140 --> 01:19:05.220 align:middle line:84% So basically, the extent that we create 01:19:05.220 --> 01:19:07.440 align:middle line:84% benefits above and beyond the wages we get, 01:19:07.440 --> 01:19:10.900 align:middle line:90% that's another argument. 01:19:10.900 --> 01:19:12.050 align:middle line:90% Other arguments? 01:19:12.050 --> 01:19:13.800 align:middle line:84% Let me just-- actually, we're out of time. 01:19:13.800 --> 01:19:16.008 align:middle line:84% So let me just hit the other two, and then we can go. 01:19:16.008 --> 01:19:21.820 align:middle line:84% The first argument is I said that high-level high school 01:19:21.820 --> 01:19:24.700 align:middle line:84% and college can't be justified by the stuff [Enoch mentioned. 01:19:24.700 --> 01:19:27.860 align:middle line:84% But what about the fact that folks 01:19:27.860 --> 01:19:31.032 align:middle line:84% need to-- who want to go to that might not have the money to? 01:19:31.032 --> 01:19:32.740 align:middle line:84% Now, the classic answer is, well, they'll 01:19:32.740 --> 01:19:34.340 align:middle line:90% just borrow from a bank. 01:19:34.340 --> 01:19:37.500 align:middle line:84% The problem is they can't because you can't borrow money 01:19:37.500 --> 01:19:39.900 align:middle line:90% from a bank without collateral. 01:19:39.900 --> 01:19:42.940 align:middle line:84% When I buy a house, I take a mortgage. 01:19:42.940 --> 01:19:44.740 align:middle line:84% The bank gives me a mortgage because they 01:19:44.740 --> 01:19:46.958 align:middle line:84% know they can take my house if I don't pay. 01:19:46.958 --> 01:19:49.000 align:middle line:84% If I take a loan from my education and don't pay, 01:19:49.000 --> 01:19:50.107 align:middle line:90% they can't take me. 01:19:50.107 --> 01:19:51.440 align:middle line:90% They can harass me on the phone. 01:19:51.440 --> 01:19:53.820 align:middle line:84% But ultimately, there's no collateral. 01:19:53.820 --> 01:19:56.688 align:middle line:84% So banks are not willing to loan for things like education 01:19:56.688 --> 01:19:58.980 align:middle line:84% because they're not sure they can get their money back, 01:19:58.980 --> 01:20:00.772 align:middle line:84% or they'll charge very high interest rates. 01:20:00.772 --> 01:20:03.360 align:middle line:84% These types of credit market failures 01:20:03.360 --> 01:20:05.760 align:middle line:84% are reasons why we might care about education. 01:20:05.760 --> 01:20:07.500 align:middle line:84% So even absent all the other things, 01:20:07.500 --> 01:20:10.040 align:middle line:84% people just can't carry out their plans. 01:20:10.040 --> 01:20:11.960 align:middle line:84% We don't want to screw-- and more 01:20:11.960 --> 01:20:15.125 align:middle line:84% generally, failures to maximize family utility. 01:20:15.125 --> 01:20:17.000 align:middle line:84% Do you really want to screw the ambitious kid 01:20:17.000 --> 01:20:20.100 align:middle line:84% with loser parents who's not willing to invest in them? 01:20:20.100 --> 01:20:22.253 align:middle line:84% No, we want what's best for society. 01:20:22.253 --> 01:20:23.920 align:middle line:84% And so once again, that's another reason 01:20:23.920 --> 01:20:25.628 align:middle line:84% why the private market might not deliver. 01:20:25.628 --> 01:20:27.720 align:middle line:84% And then finally, we have redistribution, 01:20:27.720 --> 01:20:30.680 align:middle line:84% that a great way to distribute in society 01:20:30.680 --> 01:20:32.920 align:middle line:84% is to actually bring-- is to give everyone 01:20:32.920 --> 01:20:33.987 align:middle line:90% a basic set of skills. 01:20:33.987 --> 01:20:35.820 align:middle line:84% Remember in 14.01, we talked about arguments 01:20:35.820 --> 01:20:37.200 align:middle line:90% for redistribution. 01:20:37.200 --> 01:20:40.160 align:middle line:84% And even conservatives agreed that we should at least give 01:20:40.160 --> 01:20:42.200 align:middle line:84% everyone an equal playing field, and then we 01:20:42.200 --> 01:20:43.480 align:middle line:90% should let the dice roll. 01:20:43.480 --> 01:20:45.897 align:middle line:84% But we at least want everyone have an equal playing field. 01:20:45.897 --> 01:20:49.680 align:middle line:84% That's an argument for public education's redistribution, 01:20:49.680 --> 01:20:53.230 align:middle line:84% is making sure everybody has a level playing field to start. 01:20:53.230 --> 01:20:57.000 align:middle line:90%