WEBVTT 00:00:00.000 --> 00:00:00.500 align:middle line:90% 00:00:00.500 --> 00:00:02.408 align:middle line:90% [SQUEAKING] 00:00:02.408 --> 00:00:04.316 align:middle line:90% [RUSTLING] 00:00:04.316 --> 00:00:06.701 align:middle line:90% [CLICKING] 00:00:06.701 --> 00:00:09.977 align:middle line:90% 00:00:09.977 --> 00:00:11.560 align:middle line:84% PROFESSOR: Today, we are going to turn 00:00:11.560 --> 00:00:15.120 align:middle line:84% from talking about health care to talking about inequality 00:00:15.120 --> 00:00:17.320 align:middle line:84% and government transfer programs. 00:00:17.320 --> 00:00:23.418 align:middle line:84% And I want to start this chapter with a fact 00:00:23.418 --> 00:00:25.460 align:middle line:84% that I may have referenced already in this class, 00:00:25.460 --> 00:00:28.120 align:middle line:84% but it's one of I think, the most striking facts that's 00:00:28.120 --> 00:00:33.040 align:middle line:84% emerged in public policy debates in the last several decades, 00:00:33.040 --> 00:00:38.080 align:middle line:84% which the comparison was brought to light by the Freddie Gray 00:00:38.080 --> 00:00:39.360 align:middle line:90% riots in Baltimore. 00:00:39.360 --> 00:00:42.900 align:middle line:84% In 2015, a young Black man, Freddie Gray, 00:00:42.900 --> 00:00:45.360 align:middle line:84% was arrested and killed in a police vehicle 00:00:45.360 --> 00:00:47.560 align:middle line:90% leading to a number of riots. 00:00:47.560 --> 00:00:50.200 align:middle line:84% And as people looked at the data for Baltimore, 00:00:50.200 --> 00:00:52.060 align:middle line:84% they noticed something really striking, 00:00:52.060 --> 00:00:59.300 align:middle line:84% which is shown in Figure 17.1, the first page of the handout. 00:00:59.300 --> 00:01:02.240 align:middle line:84% This shows neighborhoods in the city of Baltimore. 00:01:02.240 --> 00:01:04.780 align:middle line:90% 00:01:04.780 --> 00:01:08.500 align:middle line:84% Sandtown-Winchester is where Freddie Gray lived. 00:01:08.500 --> 00:01:12.460 align:middle line:84% Three miles away is the neighborhood of Roland Park. 00:01:12.460 --> 00:01:17.580 align:middle line:84% The average life expectancy in Roland Park is 84. 00:01:17.580 --> 00:01:19.040 align:middle line:90% That's above the US average. 00:01:19.040 --> 00:01:21.540 align:middle line:84% As you can see, the US average is about almost 79 years 00:01:21.540 --> 00:01:23.220 align:middle line:90% at this point. 00:01:23.220 --> 00:01:25.880 align:middle line:84% The average life expectancy in Sandtown-Winchester, 00:01:25.880 --> 00:01:30.460 align:middle line:84% which was three 3 apart, was 67, which was below the US life 00:01:30.460 --> 00:01:32.380 align:middle line:84% expectancy at the end of World War II 00:01:32.380 --> 00:01:34.900 align:middle line:84% and below where North Korea is today. 00:01:34.900 --> 00:01:38.700 align:middle line:84% And these are three miles apart, with a 17-year difference in how 00:01:38.700 --> 00:01:40.500 align:middle line:90% long you expect to live. 00:01:40.500 --> 00:01:42.280 align:middle line:84% This is reflected in lots of statistics. 00:01:42.280 --> 00:01:47.420 align:middle line:84% For example, the average income in Sandtown is $24,000 a year, 00:01:47.420 --> 00:01:50.500 align:middle line:84% whereas in Roland Park, it's $107,000 a year. 00:01:50.500 --> 00:01:55.980 align:middle line:84% In Roland Park, 2.5% of the children 00:01:55.980 --> 00:01:59.920 align:middle line:84% live below the federal poverty line. 00:01:59.920 --> 00:02:02.240 align:middle line:90% In Sandtown, it's 50%. 00:02:02.240 --> 00:02:06.560 align:middle line:84% So there's enormous disparities only three miles apart. 00:02:06.560 --> 00:02:09.600 align:middle line:84% And to me, that was just an incredibly striking illustration 00:02:09.600 --> 00:02:14.040 align:middle line:84% of the incredible amount of disparity we have in the US. 00:02:14.040 --> 00:02:16.880 align:middle line:84% And illustrating it really, in such an incredibly local context 00:02:16.880 --> 00:02:18.640 align:middle line:90% was powerful. 00:02:18.640 --> 00:02:22.440 align:middle line:84% Now this is important to us in public finance 00:02:22.440 --> 00:02:26.740 align:middle line:84% because part of the rationale for government intervention-- 00:02:26.740 --> 00:02:29.120 align:middle line:84% remember, I talked in the first lecture with the why 00:02:29.120 --> 00:02:31.662 align:middle line:84% should the government intervene or when should the government 00:02:31.662 --> 00:02:33.700 align:middle line:84% intervene-- one, for efficiency purpose. 00:02:33.700 --> 00:02:36.160 align:middle line:84% and that's what we focused on throughout the semester, 00:02:36.160 --> 00:02:38.720 align:middle line:84% the impact of government intervention on efficiency. 00:02:38.720 --> 00:02:40.960 align:middle line:84% The other is for redistributive purposes, 00:02:40.960 --> 00:02:44.720 align:middle line:84% that if you go back to 14.01 or Chapter 2 of the book, 00:02:44.720 --> 00:02:46.855 align:middle line:84% we have a social welfare function. 00:02:46.855 --> 00:02:49.480 align:middle line:84% That social welfare functions is where we aggregate individuals 00:02:49.480 --> 00:02:52.760 align:middle line:84% into a social measure of well-being. 00:02:52.760 --> 00:02:56.970 align:middle line:84% Most social welfare functions are implicitly redistributive. 00:02:56.970 --> 00:02:59.050 align:middle line:84% The vast majority of social welfare functions 00:02:59.050 --> 00:03:00.890 align:middle line:84% are implicitly redistributive, and that's 00:03:00.890 --> 00:03:04.530 align:middle line:84% because we have a diminishing marginal utility of consumption. 00:03:04.530 --> 00:03:06.530 align:middle line:84% So that means that the next dollar to Bill Gates 00:03:06.530 --> 00:03:08.170 align:middle line:84% means less than the next dollar to you, 00:03:08.170 --> 00:03:13.250 align:middle line:84% because he is at a vastly diminished marginal rate 00:03:13.250 --> 00:03:15.250 align:middle line:90% of consumption than you are. 00:03:15.250 --> 00:03:18.490 align:middle line:84% So as a result, most models would imply that, 00:03:18.490 --> 00:03:22.250 align:middle line:84% but redistribution is a classic public good. 00:03:22.250 --> 00:03:25.490 align:middle line:84% Unless individuals are perfectly altruistic or have a strong, 00:03:25.490 --> 00:03:30.650 align:middle line:84% warm glow, privately, they will not redistribute. 00:03:30.650 --> 00:03:32.827 align:middle line:84% Redistribution has to be mandated publicly. 00:03:32.827 --> 00:03:34.410 align:middle line:84% Privately, we'll get too little of it, 00:03:34.410 --> 00:03:36.368 align:middle line:84% by definition, like with any other public good, 00:03:36.368 --> 00:03:39.210 align:middle line:84% because redistributing is helping everyone. 00:03:39.210 --> 00:03:41.890 align:middle line:84% And so we have the classic free rider problem 00:03:41.890 --> 00:03:44.890 align:middle line:84% of redistribution, which is that basically, we'd 00:03:44.890 --> 00:03:50.690 align:middle line:84% like other people to redistribute, not us. 00:03:50.690 --> 00:03:55.118 align:middle line:84% So I'm not going to spend a lot more time on motivating why we 00:03:55.118 --> 00:03:56.410 align:middle line:90% want to redistribute in the US. 00:03:56.410 --> 00:03:58.370 align:middle line:84% I think you can go back and look at Chapter 2, 00:03:58.370 --> 00:04:00.790 align:middle line:84% or you can go back at 14.01 notes. 00:04:00.790 --> 00:04:04.070 align:middle line:84% I think we take it as a given that for most people's 00:04:04.070 --> 00:04:07.970 align:middle line:84% preferences, not everyone, but for most people preferences, 00:04:07.970 --> 00:04:10.470 align:middle line:84% there's a feeling that there is some role 00:04:10.470 --> 00:04:13.030 align:middle line:84% to play in redistributing the resources in society 00:04:13.030 --> 00:04:14.870 align:middle line:84% from the wealthiest to the least wealthy. 00:04:14.870 --> 00:04:18.855 align:middle line:84% I think that's a fairly common view. 00:04:18.855 --> 00:04:19.730 align:middle line:90% Let me rephrase that. 00:04:19.730 --> 00:04:23.470 align:middle line:84% There's a fairly common view that it is unfair, 00:04:23.470 --> 00:04:26.110 align:middle line:84% that the wealthy are so much more than the less wealthy. 00:04:26.110 --> 00:04:28.235 align:middle line:84% Whether the government should do something about it 00:04:28.235 --> 00:04:30.550 align:middle line:84% is more controversial, but it's a fairly common view 00:04:30.550 --> 00:04:31.650 align:middle line:90% that that's unfair. 00:04:31.650 --> 00:04:34.350 align:middle line:90% 00:04:34.350 --> 00:04:36.090 align:middle line:84% But to inform that view, we need data. 00:04:36.090 --> 00:04:38.670 align:middle line:84% We can't just have one comparison of two neighborhoods. 00:04:38.670 --> 00:04:43.670 align:middle line:84% We need data on comparing how unequal our society is. 00:04:43.670 --> 00:04:45.077 align:middle line:84% And so what we do in this lecture 00:04:45.077 --> 00:04:46.410 align:middle line:90% is start with some of that data. 00:04:46.410 --> 00:04:49.670 align:middle line:90% So let's go to Table 17.1. 00:04:49.670 --> 00:04:51.490 align:middle line:90% Table 17.1. 00:04:51.490 --> 00:04:57.290 align:middle line:84% shows the distribution of income in our society. 00:04:57.290 --> 00:04:59.312 align:middle line:84% And it shows it by income quintile. 00:04:59.312 --> 00:05:00.770 align:middle line:84% The way to read this table is if we 00:05:00.770 --> 00:05:05.290 align:middle line:84% had a perfectly equal society, all those numbers would be 20. 00:05:05.290 --> 00:05:07.350 align:middle line:84% Each row is 20% of the population. 00:05:07.350 --> 00:05:09.810 align:middle line:84% So it's the poorest, 20%, the next 20%, et cetera, all 00:05:09.810 --> 00:05:11.570 align:middle line:90% up to the richest 20%. 00:05:11.570 --> 00:05:12.990 align:middle line:84% So if society is perfectly equal, 00:05:12.990 --> 00:05:14.770 align:middle line:90% all those numbers would be 20. 00:05:14.770 --> 00:05:18.610 align:middle line:84% If you look at 1967, that wasn't true. 00:05:18.610 --> 00:05:22.370 align:middle line:84% The 20% of the poorest people in society had only 4% 00:05:22.370 --> 00:05:26.090 align:middle line:84% of the income, whereas the richest 20% had 10 times 00:05:26.090 --> 00:05:26.770 align:middle line:90% as much. 00:05:26.770 --> 00:05:29.790 align:middle line:90% 00:05:29.790 --> 00:05:34.290 align:middle line:84% That gap shrunk initially before rising dramatically again, 00:05:34.290 --> 00:05:36.890 align:middle line:84% to the point where today, the poorest 00:05:36.890 --> 00:05:39.530 align:middle line:84% have much less than they did in 1967, 00:05:39.530 --> 00:05:42.165 align:middle line:84% and the richest control more than half the resources. 00:05:42.165 --> 00:05:43.790 align:middle line:84% So now, instead of that ratio being 10, 00:05:43.790 --> 00:05:48.250 align:middle line:90% it's more like 16 or 17. 00:05:48.250 --> 00:05:51.350 align:middle line:84% So basically, we've become much more unequal 00:05:51.350 --> 00:05:53.108 align:middle line:90% as a society over time. 00:05:53.108 --> 00:05:55.150 align:middle line:84% One of the other striking ways to illustrate that 00:05:55.150 --> 00:05:57.710 align:middle line:84% is to look at Figure 17.2, which highlights 00:05:57.710 --> 00:05:59.590 align:middle line:84% that much of the growth of inequality 00:05:59.590 --> 00:06:01.250 align:middle line:90% is at the very, very top. 00:06:01.250 --> 00:06:02.750 align:middle line:84% This shows the share of income going 00:06:02.750 --> 00:06:06.030 align:middle line:90% to the top 1% of Americans. 00:06:06.030 --> 00:06:11.790 align:middle line:84% Today, one in every $4 goes to the 1% richest Americans, 00:06:11.790 --> 00:06:14.290 align:middle line:84% which was also true at the beginning of the 20th century, 00:06:14.290 --> 00:06:15.830 align:middle line:90% actually, somewhat lower. 00:06:15.830 --> 00:06:18.910 align:middle line:84% It then fell in the middle of the 20th century, 00:06:18.910 --> 00:06:23.210 align:middle line:84% before rising again in the 21st century, 00:06:23.210 --> 00:06:25.270 align:middle line:84% in the end of the 20th century, the 21st century, 00:06:25.270 --> 00:06:26.730 align:middle line:90% to be much higher levels. 00:06:26.730 --> 00:06:27.710 align:middle line:90% Yeah? 00:06:27.710 --> 00:06:30.310 align:middle line:84% AUDIENCE: How much more skew [INAUDIBLE]? 00:06:30.310 --> 00:06:31.870 align:middle line:90% PROFESSOR: Much more skewed. 00:06:31.870 --> 00:06:33.533 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:06:33.533 --> 00:06:35.450 align:middle line:84% PROFESSOR: The rich would have a larger share. 00:06:35.450 --> 00:06:38.330 align:middle line:84% So the richest 1% have about 25% of the income. 00:06:38.330 --> 00:06:41.990 align:middle line:84% They have probably at least 50% of the wealth. 00:06:41.990 --> 00:06:43.730 align:middle line:84% And we'll talk about why that is. 00:06:43.730 --> 00:06:46.090 align:middle line:84% The bottom line is because the wealthier you are, 00:06:46.090 --> 00:06:48.240 align:middle line:90% the more you can save. 00:06:48.240 --> 00:06:51.187 align:middle line:84% So basically, there's enormous inequality. 00:06:51.187 --> 00:06:53.520 align:middle line:84% You can actually look at us compared to other countries, 00:06:53.520 --> 00:06:54.793 align:middle line:90% because what's unequal? 00:06:54.793 --> 00:06:55.460 align:middle line:90% What defines it? 00:06:55.460 --> 00:06:58.520 align:middle line:84% Well, when we look at other countries-- look at Table 17.2. 00:06:58.520 --> 00:07:00.780 align:middle line:84% This shows slightly different numbers. 00:07:00.780 --> 00:07:03.460 align:middle line:84% This shows different parts of the income distribution, 00:07:03.460 --> 00:07:07.920 align:middle line:84% so the bottom 10% all the way up to the top 10% and the middle 00:07:07.920 --> 00:07:11.640 align:middle line:84% as well across developed countries. 00:07:11.640 --> 00:07:15.280 align:middle line:84% And what you see is the US has the lowest 00:07:15.280 --> 00:07:19.040 align:middle line:84% share of its population in the bottom 10% 00:07:19.040 --> 00:07:24.340 align:middle line:84% of any other developed country on this list, and the US has-- 00:07:24.340 --> 00:07:27.480 align:middle line:84% I think it's the second highest share 00:07:27.480 --> 00:07:32.740 align:middle line:84% of its income in the top 10%, other than Mexico. 00:07:32.740 --> 00:07:35.935 align:middle line:84% So no country has a lower share of its income going to the poor, 00:07:35.935 --> 00:07:37.560 align:middle line:84% and only one country has a higher share 00:07:37.560 --> 00:07:41.080 align:middle line:84% of its income going to the wealthy than does the US. 00:07:41.080 --> 00:07:44.760 align:middle line:84% So unequal both historically and relative to the rest 00:07:44.760 --> 00:07:47.220 align:middle line:84% of the world, we have extreme inequality. 00:07:47.220 --> 00:07:48.740 align:middle line:84% And that's what motivates the notion 00:07:48.740 --> 00:07:52.500 align:middle line:84% that we need to take some action to address that inequality. 00:07:52.500 --> 00:07:55.180 align:middle line:84% It motivates the notion that it's unfair 00:07:55.180 --> 00:07:57.640 align:middle line:84% that societies distribute resources this way, 00:07:57.640 --> 00:07:59.640 align:middle line:84% and that there may be a role for the government, 00:07:59.640 --> 00:08:04.420 align:middle line:84% may be a role for the government in addressing that inequality. 00:08:04.420 --> 00:08:09.320 align:middle line:84% Now that is one way to think about society's inequality, 00:08:09.320 --> 00:08:11.260 align:middle line:84% but another way is say, wait a second. 00:08:11.260 --> 00:08:13.100 align:middle line:90% Why do I care? 00:08:13.100 --> 00:08:15.960 align:middle line:84% Maybe all I care about is not this relative measure, 00:08:15.960 --> 00:08:19.780 align:middle line:84% but a measure of absolute deprivation. 00:08:19.780 --> 00:08:24.620 align:middle line:84% That is, imagine a world where everyone had enough to eat, 00:08:24.620 --> 00:08:27.340 align:middle line:90% housing, clothes, whatever. 00:08:27.340 --> 00:08:29.860 align:middle line:84% And in that world, someone got really rich. 00:08:29.860 --> 00:08:31.620 align:middle line:90% Would we care so much? 00:08:31.620 --> 00:08:33.500 align:middle line:84% I mean, as long as everyone's doing OK, 00:08:33.500 --> 00:08:36.039 align:middle line:84% why do I really care how much money rich people have? 00:08:36.039 --> 00:08:37.900 align:middle line:84% That is, why do we care about the relative? 00:08:37.900 --> 00:08:39.760 align:middle line:84% It really is the absolute that matters. 00:08:39.760 --> 00:08:41.177 align:middle line:84% It's making sure that everyone can 00:08:41.177 --> 00:08:43.860 align:middle line:84% live a decent standard of living that really matters, 00:08:43.860 --> 00:08:46.400 align:middle line:84% and that's led to a focus, instead of on instead 00:08:46.400 --> 00:08:48.360 align:middle line:84% of relative measures of inequality, 00:08:48.360 --> 00:08:50.880 align:middle line:84% on absolute measures of inequality. 00:08:50.880 --> 00:08:54.210 align:middle line:84% And the most famous of those is the poverty line. 00:08:54.210 --> 00:08:55.960 align:middle line:84% This is a great story, actually, for those 00:08:55.960 --> 00:08:59.400 align:middle line:84% who believe in the power of government bureaucrats 00:08:59.400 --> 00:09:01.490 align:middle line:90% to do good and powerful things. 00:09:01.490 --> 00:09:03.240 align:middle line:84% There's a woman named Mollie Orshansky who 00:09:03.240 --> 00:09:05.573 align:middle line:84% was working in the Division of Health and Human Services 00:09:05.573 --> 00:09:06.073 align:middle line:90% in 1964. 00:09:06.073 --> 00:09:08.531 align:middle line:84% It was actually called the Department of Health, Education, 00:09:08.531 --> 00:09:09.760 align:middle line:90% and Welfare then. 00:09:09.760 --> 00:09:12.960 align:middle line:84% And she was charged with the exercise of figuring out what 00:09:12.960 --> 00:09:16.313 align:middle line:90% it means to be poor in America. 00:09:16.313 --> 00:09:17.480 align:middle line:90% She went to the [INAUDIBLE]. 00:09:17.480 --> 00:09:19.730 align:middle line:84% She said people in America, on average, at that point, 00:09:19.730 --> 00:09:21.660 align:middle line:84% spend a third of their budget on food. 00:09:21.660 --> 00:09:23.160 align:middle line:84% So she said, well, what does it cost 00:09:23.160 --> 00:09:26.080 align:middle line:84% to have a minimally, nutritionally adequate diet? 00:09:26.080 --> 00:09:29.420 align:middle line:84% She calculated that, she multiplied it by 3, and said, 00:09:29.420 --> 00:09:31.760 align:middle line:84% this is what it costs to live in America. 00:09:31.760 --> 00:09:34.960 align:middle line:84% That calculation is, today, the basis for trillions 00:09:34.960 --> 00:09:36.580 align:middle line:90% of dollars of redistribution. 00:09:36.580 --> 00:09:39.320 align:middle line:84% We've taken Mollie Orshansky calculation and just updated it 00:09:39.320 --> 00:09:40.713 align:middle line:90% by inflation ever since. 00:09:40.713 --> 00:09:42.380 align:middle line:84% And the poverty line, which I've already 00:09:42.380 --> 00:09:44.380 align:middle line:84% mentioned a number of times without defining it, 00:09:44.380 --> 00:09:47.340 align:middle line:84% is the basis for trillions of dollars of the redistribution we 00:09:47.340 --> 00:09:49.740 align:middle line:90% do in our society. 00:09:49.740 --> 00:09:52.500 align:middle line:84% So the poverty line is our classic absolute deprivation 00:09:52.500 --> 00:09:53.000 align:middle line:90% measure. 00:09:53.000 --> 00:09:57.593 align:middle line:84% You can see what the poverty line looks like in Table 17.3. 00:09:57.593 --> 00:09:58.760 align:middle line:90% I've been highlighting this. 00:09:58.760 --> 00:10:00.380 align:middle line:84% I've been talking about this before. 00:10:00.380 --> 00:10:03.540 align:middle line:84% Basically, the way this works is, 00:10:03.540 --> 00:10:05.780 align:middle line:84% it doesn't go up multiplicatively 00:10:05.780 --> 00:10:07.020 align:middle line:90% with family size. 00:10:07.020 --> 00:10:09.020 align:middle line:84% We assume that there's economies of scale 00:10:09.020 --> 00:10:10.620 align:middle line:90% of people living together. 00:10:10.620 --> 00:10:14.360 align:middle line:84% So the poverty line for one individual is $13,000. 00:10:14.360 --> 00:10:16.268 align:middle line:84% For two people, it's not twice as much, 00:10:16.268 --> 00:10:18.560 align:middle line:84% because we think there's econ-- you can share the rent, 00:10:18.560 --> 00:10:21.015 align:middle line:84% you can share food, and you can share cleaning, cetera. 00:10:21.015 --> 00:10:23.140 align:middle line:84% So basically, the poverty line for a family of four 00:10:23.140 --> 00:10:26.780 align:middle line:84% is about twice the poverty line for an individual. 00:10:26.780 --> 00:10:29.205 align:middle line:90% So those are the poverty lines. 00:10:29.205 --> 00:10:30.580 align:middle line:84% Now you can look at those numbers 00:10:30.580 --> 00:10:31.900 align:middle line:90% and make of them what you will. 00:10:31.900 --> 00:10:33.400 align:middle line:84% If you're from an affluent part of the country, 00:10:33.400 --> 00:10:34.700 align:middle line:84% you'd be like, oh my god, how can anyone live 00:10:34.700 --> 00:10:35.638 align:middle line:90% on anything near this? 00:10:35.638 --> 00:10:37.680 align:middle line:84% If you're in a less affluent part of the country, 00:10:37.680 --> 00:10:40.060 align:middle line:84% maybe those don't look so crazy to you, 00:10:40.060 --> 00:10:44.942 align:middle line:84% but that's our measure of the minimum you need to survive. 00:10:44.942 --> 00:10:46.400 align:middle line:84% Now what's happened to that measure 00:10:46.400 --> 00:10:47.775 align:middle line:84% is quite different than equality. 00:10:47.775 --> 00:10:51.465 align:middle line:84% If you look at figure 17.3, you see a somewhat different pattern 00:10:51.465 --> 00:10:52.840 align:middle line:84% than we saw for inequality, which 00:10:52.840 --> 00:10:56.760 align:middle line:84% is if you look at everyone, all people, that's the red line. 00:10:56.760 --> 00:11:00.520 align:middle line:84% You see a dramatic decline in poverty during the 1960s, 00:11:00.520 --> 00:11:03.440 align:middle line:84% and it's basically been flat since. 00:11:03.440 --> 00:11:05.220 align:middle line:84% Now what's interesting about the decline, 00:11:05.220 --> 00:11:07.240 align:middle line:84% it really was driven by the elderly. 00:11:07.240 --> 00:11:10.620 align:middle line:84% If you look at the 65 and over, that's the blue line. 00:11:10.620 --> 00:11:12.902 align:middle line:84% They had the dramatic reduction in poverty, 00:11:12.902 --> 00:11:15.360 align:middle line:84% whereas those, for example, for kids, the poverty reduction 00:11:15.360 --> 00:11:16.760 align:middle line:90% wasn't so dramatic. 00:11:16.760 --> 00:11:19.720 align:middle line:84% But the key thing is if you look over the last few decades, 00:11:19.720 --> 00:11:22.300 align:middle line:90% it's pretty flat. 00:11:22.300 --> 00:11:25.560 align:middle line:84% We saw this enormous increase in income inequality. 00:11:25.560 --> 00:11:27.520 align:middle line:84% You don't see an enormous increase in poverty. 00:11:27.520 --> 00:11:29.320 align:middle line:90% It looks pretty flat. 00:11:29.320 --> 00:11:32.080 align:middle line:84% So that says that basically, how you 00:11:32.080 --> 00:11:34.742 align:middle line:84% feel about what's happening to resource in society 00:11:34.742 --> 00:11:37.200 align:middle line:84% and the sharing the pie depends on how you think about what 00:11:37.200 --> 00:11:39.050 align:middle line:90% the right measure is. 00:11:39.050 --> 00:11:44.010 align:middle line:84% Now, with that in mind, there are two problems. 00:11:44.010 --> 00:11:47.250 align:middle line:84% There are a few problems with relying 00:11:47.250 --> 00:11:49.290 align:middle line:90% in this absolute measure. 00:11:49.290 --> 00:11:52.050 align:middle line:84% The first problem is, you have to decide what the level is. 00:11:52.050 --> 00:11:54.850 align:middle line:84% And while Mollie Orshansky did an admirable job, 00:11:54.850 --> 00:11:57.330 align:middle line:84% her answer is increasingly wrong. 00:11:57.330 --> 00:11:59.390 align:middle line:84% It's increasingly wrong for a number of reasons. 00:11:59.390 --> 00:12:02.610 align:middle line:84% First of all, food is no longer about one-third 00:12:02.610 --> 00:12:04.750 align:middle line:84% of the typical household's budget. 00:12:04.750 --> 00:12:08.290 align:middle line:84% It's now 13% of the typical household's budget, 00:12:08.290 --> 00:12:11.425 align:middle line:84% and inflation and food has been much lower than inflation 00:12:11.425 --> 00:12:14.050 align:middle line:84% in other goods, in particularly housing and medical care, which 00:12:14.050 --> 00:12:16.890 align:middle line:84% have grown to be a much larger share of a person's budget. 00:12:16.890 --> 00:12:20.850 align:middle line:84% So in fact, living at the same level as 1964 00:12:20.850 --> 00:12:23.330 align:middle line:84% has gotten much harder, because the price of things 00:12:23.330 --> 00:12:25.330 align:middle line:84% that are a bigger and bigger share of our budget 00:12:25.330 --> 00:12:27.090 align:middle line:84% are the ones that are rising the fastest. 00:12:27.090 --> 00:12:28.383 align:middle line:90% That's one problem. 00:12:28.383 --> 00:12:30.050 align:middle line:84% The second problem with the poverty line 00:12:30.050 --> 00:12:32.250 align:middle line:84% is, it's one national line, whereas we 00:12:32.250 --> 00:12:35.530 align:middle line:84% know the cost of living differs enormously around the country. 00:12:35.530 --> 00:12:37.030 align:middle line:84% So that's another problem with this, 00:12:37.030 --> 00:12:40.150 align:middle line:84% is you really like something which varies locally. 00:12:40.150 --> 00:12:41.990 align:middle line:84% And then the third conceptual problem 00:12:41.990 --> 00:12:44.290 align:middle line:84% is that it's a very incomplete measure of income. 00:12:44.290 --> 00:12:47.070 align:middle line:84% For example, this is people's cash income. 00:12:47.070 --> 00:12:51.050 align:middle line:84% If you get Medicaid, it doesn't count as a resource for you. 00:12:51.050 --> 00:12:52.858 align:middle line:84% So if I take two people, if I'm Matt 00:12:52.858 --> 00:12:55.150 align:middle line:84% and I have the same cash income, but Matt gets Medicaid 00:12:55.150 --> 00:12:57.670 align:middle line:84% and I don't, we're considered equally poor. 00:12:57.670 --> 00:12:59.730 align:middle line:84% Well, no, but Matt's got something I don't have. 00:12:59.730 --> 00:13:00.870 align:middle line:90% He has Medicaid. 00:13:00.870 --> 00:13:02.550 align:middle line:90% That's not counted. 00:13:02.550 --> 00:13:03.530 align:middle line:90% It's pre-tax. 00:13:03.530 --> 00:13:05.050 align:middle line:84% So if I pay bigger taxes than Matt, 00:13:05.050 --> 00:13:07.710 align:middle line:84% so therefore, I have fewer resources, that's not counted. 00:13:07.710 --> 00:13:09.590 align:middle line:84% So there's a number of incomplete features 00:13:09.590 --> 00:13:11.590 align:middle line:90% of this system. 00:13:11.590 --> 00:13:14.190 align:middle line:84% Now this has been known for a long time, 00:13:14.190 --> 00:13:16.650 align:middle line:84% and there was an effort to try to revise the poverty line. 00:13:16.650 --> 00:13:19.067 align:middle line:84% So let me tell you one of the most interesting things that 00:13:19.067 --> 00:13:20.310 align:middle line:90% happened to me in Washington. 00:13:20.310 --> 00:13:22.843 align:middle line:84% I had a terrific secretary who was great. 00:13:22.843 --> 00:13:24.010 align:middle line:90% She took care of everything. 00:13:24.010 --> 00:13:24.770 align:middle line:90% She knew all the meetings. 00:13:24.770 --> 00:13:25.570 align:middle line:90% She said, go to that one. 00:13:25.570 --> 00:13:26.090 align:middle line:90% That's the big meeting. 00:13:26.090 --> 00:13:26.730 align:middle line:90% That's not a good one. 00:13:26.730 --> 00:13:27.150 align:middle line:90% Whatever. 00:13:27.150 --> 00:13:28.010 align:middle line:90% One day, she comes to me. 00:13:28.010 --> 00:13:30.130 align:middle line:84% She says, John, this has never happened before, 00:13:30.130 --> 00:13:31.910 align:middle line:84% but they have a meeting on your books 00:13:31.910 --> 00:13:34.147 align:middle line:84% and they won't tell me where it is and what it's for. 00:13:34.147 --> 00:13:35.730 align:middle line:84% She said it has never happened before, 00:13:35.730 --> 00:13:36.870 align:middle line:90% in 30 years in government. 00:13:36.870 --> 00:13:38.290 align:middle line:84% She said, I don't know if we're at war 00:13:38.290 --> 00:13:40.650 align:middle line:84% or what the hell's going on, but this meeting is in an hour, 00:13:40.650 --> 00:13:42.970 align:middle line:84% and they won't tell me what it's about or where it is. 00:13:42.970 --> 00:13:44.630 align:middle line:84% Someone's going to come get you and bring you there. 00:13:44.630 --> 00:13:46.837 align:middle line:84% I'm like, wow, this could be pretty serious. 00:13:46.837 --> 00:13:47.670 align:middle line:90% So they come get me. 00:13:47.670 --> 00:13:50.102 align:middle line:84% They bring me down to this soundproof room, 00:13:50.102 --> 00:13:50.810 align:middle line:90% and we sit there. 00:13:50.810 --> 00:13:54.170 align:middle line:84% And they say, we're thinking about revising the poverty line. 00:13:54.170 --> 00:13:55.728 align:middle line:90% And I'm like, I'm sorry? 00:13:55.728 --> 00:13:57.770 align:middle line:84% And they said, well, as long as the poverty line. 00:13:57.770 --> 00:13:59.190 align:middle line:84% We want to revise it and we're thinking about it. 00:13:59.190 --> 00:14:00.480 align:middle line:90% Why the secret room? 00:14:00.480 --> 00:14:02.730 align:middle line:84% Well, because revising the poverty line is politically 00:14:02.730 --> 00:14:04.990 align:middle line:84% nightmarish, because for example, 00:14:04.990 --> 00:14:07.870 align:middle line:84% if you revise the poverty line to include benefits, 00:14:07.870 --> 00:14:10.430 align:middle line:84% that's going to make a lot of Black people seem less poor 00:14:10.430 --> 00:14:11.865 align:middle line:90% and white people seem more poor. 00:14:11.865 --> 00:14:13.990 align:middle line:84% So it's going to be a big redistribution of wealth. 00:14:13.990 --> 00:14:15.550 align:middle line:84% We give money based on poverty line. 00:14:15.550 --> 00:14:16.710 align:middle line:84% If we revise to include benefits, 00:14:16.710 --> 00:14:18.250 align:middle line:84% suddenly we're shifting a bunch of money from Black people 00:14:18.250 --> 00:14:18.990 align:middle line:90% to white people. 00:14:18.990 --> 00:14:22.600 align:middle line:84% We're shifting a ton of money from Mississippi to New York, 00:14:22.600 --> 00:14:24.850 align:middle line:84% because people in New York who are at the poverty line 00:14:24.850 --> 00:14:27.050 align:middle line:84% are way poorer, effectively, than people in Mississippi. 00:14:27.050 --> 00:14:29.150 align:middle line:84% The political implications would have been a nightmare. 00:14:29.150 --> 00:14:30.817 align:middle line:84% And that's why they keep it super secret 00:14:30.817 --> 00:14:32.330 align:middle line:90% and decided never to do it. 00:14:32.330 --> 00:14:34.350 align:middle line:84% There is something called a Supplemental Poverty 00:14:34.350 --> 00:14:36.670 align:middle line:84% Measure, which was then introduced, 00:14:36.670 --> 00:14:39.930 align:middle line:84% which you can track, and is a better measure of poverty, 00:14:39.930 --> 00:14:42.430 align:middle line:84% but it's never been adopted or an official government policy 00:14:42.430 --> 00:14:43.850 align:middle line:90% because it's too controversial. 00:14:43.850 --> 00:14:44.470 align:middle line:90% Yeah? 00:14:44.470 --> 00:14:49.692 align:middle line:84% AUDIENCE: Is that [INAUDIBLE] more about because the people 00:14:49.692 --> 00:14:51.650 align:middle line:84% are getting from the government or because of-- 00:14:51.650 --> 00:14:53.150 align:middle line:84% PROFESSOR: No, it's just cost of living. 00:14:53.150 --> 00:14:53.730 align:middle line:90% The cost of living. 00:14:53.730 --> 00:14:54.930 align:middle line:90% They made one poverty line. 00:14:54.930 --> 00:14:56.610 align:middle line:84% So everybody below the poverty line got money. 00:14:56.610 --> 00:14:57.870 align:middle line:84% Well, shit, that was half the people in Mississippi 00:14:57.870 --> 00:14:59.348 align:middle line:90% and 10% of people in New York. 00:14:59.348 --> 00:15:01.890 align:middle line:84% So if you had poverty line that reflected the cost of living, 00:15:01.890 --> 00:15:03.260 align:middle line:84% you'd be giving a lot less to Mississippi 00:15:03.260 --> 00:15:04.450 align:middle line:90% and a lot more to New York. 00:15:04.450 --> 00:15:06.390 align:middle line:84% That was politically problematic. 00:15:06.390 --> 00:15:09.850 align:middle line:84% So one problem is, you've got to define it, 00:15:09.850 --> 00:15:12.350 align:middle line:90% and that's incredibly hard. 00:15:12.350 --> 00:15:16.910 align:middle line:84% The second problem is that what it is, 00:15:16.910 --> 00:15:18.410 align:middle line:84% not only have to define technically, 00:15:18.410 --> 00:15:21.210 align:middle line:84% but what does it even mean to live minimally, 00:15:21.210 --> 00:15:22.470 align:middle line:90% adequately in society? 00:15:22.470 --> 00:15:25.550 align:middle line:84% So in 1950, you didn't have to have a car 00:15:25.550 --> 00:15:26.930 align:middle line:90% to live adequately in society. 00:15:26.930 --> 00:15:28.310 align:middle line:90% Today, you do. 00:15:28.310 --> 00:15:31.400 align:middle line:84% So just even what it means is really hard. 00:15:31.400 --> 00:15:33.900 align:middle line:84% It's like the Inflation debate you've talked about in macro. 00:15:33.900 --> 00:15:38.680 align:middle line:84% You have to really define this, redefine this over time. 00:15:38.680 --> 00:15:40.640 align:middle line:90% So that's one problem. 00:15:40.640 --> 00:15:44.880 align:middle line:84% The other problem we face is that it 00:15:44.880 --> 00:15:48.320 align:middle line:84% may be that actually, relative income inequality actually 00:15:48.320 --> 00:15:49.800 align:middle line:90% matters. 00:15:49.800 --> 00:15:52.680 align:middle line:84% It may be that it turns out that actually, we care not just 00:15:52.680 --> 00:15:55.080 align:middle line:84% about how much the poor have, but how much the rich have 00:15:55.080 --> 00:15:56.040 align:middle line:90% as well. 00:15:56.040 --> 00:15:59.600 align:middle line:84% And this comes from basic biology. 00:15:59.600 --> 00:16:03.120 align:middle line:84% So they ran an experiment with a set of monkeys, 00:16:03.120 --> 00:16:06.400 align:middle line:84% and the monkeys were trained to hand over pebbles in exchange 00:16:06.400 --> 00:16:09.080 align:middle line:84% for cucumber slices, which they liked. 00:16:09.080 --> 00:16:13.120 align:middle line:84% But then one monkey was often offered a better deal. 00:16:13.120 --> 00:16:15.620 align:middle line:84% If he gave a pebble, he didn't get a cucumber slice. 00:16:15.620 --> 00:16:18.920 align:middle line:84% He got a grape, which the monkeys liked even more. 00:16:18.920 --> 00:16:21.360 align:middle line:84% But then the second monkey was then offered the same deal 00:16:21.360 --> 00:16:22.982 align:middle line:90% as before, and it revolted. 00:16:22.982 --> 00:16:23.940 align:middle line:90% It said, wait a second. 00:16:23.940 --> 00:16:24.740 align:middle line:90% That's crazy. 00:16:24.740 --> 00:16:25.960 align:middle line:84% That monkey got offered a grape, but I'm just 00:16:25.960 --> 00:16:26.940 align:middle line:90% getting a cucumber slice. 00:16:26.940 --> 00:16:28.420 align:middle line:84% And you couldn't reason with the monkey, 00:16:28.420 --> 00:16:29.460 align:middle line:84% but you'd like to say, well, look, it's 00:16:29.460 --> 00:16:30.400 align:middle line:90% the same deal you had before. 00:16:30.400 --> 00:16:31.200 align:middle line:90% What's the problem? 00:16:31.200 --> 00:16:33.780 align:middle line:84% The problem was, someone else was getting a better deal. 00:16:33.780 --> 00:16:37.620 align:middle line:84% And that's actually the way our utility functions operate. 00:16:37.620 --> 00:16:39.660 align:middle line:84% If you think about it, if we define 00:16:39.660 --> 00:16:44.680 align:middle line:84% our happiness and our sense of well-being, it's all relative. 00:16:44.680 --> 00:16:47.363 align:middle line:84% You talk about how, in fact, people in very poor societies 00:16:47.363 --> 00:16:49.280 align:middle line:84% are equally happy as people in rich societies. 00:16:49.280 --> 00:16:51.697 align:middle line:84% It's not quite true, but it's closer than you might think, 00:16:51.697 --> 00:16:53.660 align:middle line:84% because everyone around them is poor. 00:16:53.660 --> 00:16:55.700 align:middle line:84% And in fact, there's lots of evidence 00:16:55.700 --> 00:16:57.220 align:middle line:84% that basically, people's well-being 00:16:57.220 --> 00:17:00.500 align:middle line:84% is about relative income, not absolute income. 00:17:00.500 --> 00:17:05.780 align:middle line:84% So for example, if you look at people's reported self 00:17:05.780 --> 00:17:08.500 align:middle line:84% well-being, it depends more on their neighbor's income 00:17:08.500 --> 00:17:09.480 align:middle line:90% than their income. 00:17:09.480 --> 00:17:14.540 align:middle line:90% 00:17:14.540 --> 00:17:16.492 align:middle line:84% They did a cash transfer program in Kenya, 00:17:16.492 --> 00:17:18.200 align:middle line:84% where they gave people money, people that 00:17:18.200 --> 00:17:19.599 align:middle line:90% got the money were happier. 00:17:19.599 --> 00:17:21.266 align:middle line:84% The people whose neighbors got the money 00:17:21.266 --> 00:17:23.540 align:middle line:84% were sadder, regardless of whether they 00:17:23.540 --> 00:17:25.140 align:middle line:90% got the money or not. 00:17:25.140 --> 00:17:32.000 align:middle line:84% And then one of my favorite studies of this 00:17:32.000 --> 00:17:34.080 align:middle line:84% is that basically, lottery winners 00:17:34.080 --> 00:17:39.340 align:middle line:84% in Canada that-- when someone won the lottery in Canada, 00:17:39.340 --> 00:17:42.370 align:middle line:84% their neighbors were more likely to go bankrupt. 00:17:42.370 --> 00:17:43.620 align:middle line:90% Think about that for a second. 00:17:43.620 --> 00:17:45.080 align:middle line:84% When somebody won the lottery, their neighbors 00:17:45.080 --> 00:17:47.240 align:middle line:84% were more likely to go bankrupt, because they suddenly 00:17:47.240 --> 00:17:48.940 align:middle line:84% had to spend more to keep up with their friends. 00:17:48.940 --> 00:17:50.440 align:middle line:84% Their friend won the lottery, so they're like, 00:17:50.440 --> 00:17:51.440 align:middle line:90% oh, he's got a nice car. 00:17:51.440 --> 00:17:52.600 align:middle line:90% I got a nice car. 00:17:52.600 --> 00:17:54.280 align:middle line:84% So basically, the point is we all 00:17:54.280 --> 00:17:57.893 align:middle line:84% derive our sense of what's fair and what's just 00:17:57.893 --> 00:17:59.060 align:middle line:90% through relative inequality. 00:17:59.060 --> 00:18:00.520 align:middle line:84% So it's not obvious that relevant inequality 00:18:00.520 --> 00:18:02.228 align:middle line:84% is the wrong measure to look at, but it's 00:18:02.228 --> 00:18:04.320 align:middle line:84% a very deep and hard debate, which 00:18:04.320 --> 00:18:06.640 align:middle line:84% is should we care how much Elon Musk, to put 00:18:06.640 --> 00:18:10.040 align:middle line:90% an evil name to the story? 00:18:10.040 --> 00:18:12.260 align:middle line:84% Should we really care how much he has? 00:18:12.260 --> 00:18:14.500 align:middle line:90% At one level, we shouldn't. 00:18:14.500 --> 00:18:17.220 align:middle line:84% He's invented some incredibly cool shit. 00:18:17.220 --> 00:18:21.480 align:middle line:84% Let him have as much money as he wants, as long as the rest of us 00:18:21.480 --> 00:18:22.640 align:middle line:90% have enough to get by. 00:18:22.640 --> 00:18:25.517 align:middle line:84% At another level, if our utility is really 00:18:25.517 --> 00:18:27.100 align:middle line:84% about relative inequality and the role 00:18:27.100 --> 00:18:29.420 align:middle line:84% the government has to maximize social welfare, 00:18:29.420 --> 00:18:30.620 align:middle line:90% then we should care. 00:18:30.620 --> 00:18:33.140 align:middle line:90% And that's the real challenge. 00:18:33.140 --> 00:18:36.060 align:middle line:90% Questions about that? 00:18:36.060 --> 00:18:39.180 align:middle line:84% So regardless of what the right measure is, the bottom line is, 00:18:39.180 --> 00:18:41.380 align:middle line:90% people care about unfairness. 00:18:41.380 --> 00:18:44.022 align:middle line:84% And then the question is, well, what can the government do? 00:18:44.022 --> 00:18:45.980 align:middle line:84% So the first question is, should the government 00:18:45.980 --> 00:18:46.680 align:middle line:90% do something about it? 00:18:46.680 --> 00:18:48.972 align:middle line:84% And the answer is, well, there's an obvious theoretical 00:18:48.972 --> 00:18:52.515 align:middle line:84% justification, which is redistribution is a public good. 00:18:52.515 --> 00:18:54.140 align:middle line:84% There's charity, but the private sector 00:18:54.140 --> 00:18:56.220 align:middle line:84% will not do the right amount of redistribution. 00:18:56.220 --> 00:18:58.260 align:middle line:90% It's a public good. 00:18:58.260 --> 00:18:59.720 align:middle line:84% So what should the government do? 00:18:59.720 --> 00:19:01.137 align:middle line:84% Well, what the government does is, 00:19:01.137 --> 00:19:03.780 align:middle line:84% it has a number of what we call transfer programs. 00:19:03.780 --> 00:19:06.540 align:middle line:84% And these transfer programs are distinguished in two 00:19:06.540 --> 00:19:09.340 align:middle line:84% very important ways that we're going to have to keep in mind. 00:19:09.340 --> 00:19:13.180 align:middle line:84% This is not Medicaid level, but you got to know this. 00:19:13.180 --> 00:19:16.340 align:middle line:84% The first way they're distinguished 00:19:16.340 --> 00:19:23.840 align:middle line:84% is categorical versus means tested. 00:19:23.840 --> 00:19:28.520 align:middle line:90% 00:19:28.520 --> 00:19:29.540 align:middle line:90% What does that mean? 00:19:29.540 --> 00:19:31.560 align:middle line:84% A categorical welfare program is one 00:19:31.560 --> 00:19:35.200 align:middle line:84% that you get based on your characteristics, like being 00:19:35.200 --> 00:19:38.000 align:middle line:90% blind, being a single mother. 00:19:38.000 --> 00:19:42.560 align:middle line:84% Means tested is something where you get purely on your income. 00:19:42.560 --> 00:19:46.060 align:middle line:84% And welfare programs sometimes can be one, sometimes the other, 00:19:46.060 --> 00:19:48.600 align:middle line:90% sometimes both. 00:19:48.600 --> 00:19:50.760 align:middle line:90% So that's the first distinction. 00:19:50.760 --> 00:19:58.120 align:middle line:84% The second distinction is cash versus in kind, 00:19:58.120 --> 00:19:59.760 align:middle line:84% which is are you giving people money, 00:19:59.760 --> 00:20:02.440 align:middle line:90% or are you giving them goods? 00:20:02.440 --> 00:20:04.222 align:middle line:84% And these are two important distinctions 00:20:04.222 --> 00:20:06.680 align:middle line:84% you want to keep in mind as we talk about different welfare 00:20:06.680 --> 00:20:07.600 align:middle line:90% programs. 00:20:07.600 --> 00:20:10.940 align:middle line:84% So let's start by talking about our big cash. 00:20:10.940 --> 00:20:13.640 align:middle line:84% And by the way, the term welfare gets 00:20:13.640 --> 00:20:16.440 align:middle line:84% hard to think about here, because we typically, 00:20:16.440 --> 00:20:18.023 align:middle line:84% in the US, when we say welfare, we 00:20:18.023 --> 00:20:19.940 align:middle line:84% don't mean the concept of economic well-being. 00:20:19.940 --> 00:20:22.530 align:middle line:84% We mean cash transfers to poor people. 00:20:22.530 --> 00:20:24.822 align:middle line:84% So the term welfare is thrown around political debates. 00:20:24.822 --> 00:20:26.572 align:middle line:84% It's not the term welfare as we've used it 00:20:26.572 --> 00:20:27.890 align:middle line:90% so far in this semester. 00:20:27.890 --> 00:20:29.650 align:middle line:90% It means redistribution. 00:20:29.650 --> 00:20:33.810 align:middle line:84% So there's two big cash transfer programs. 00:20:33.810 --> 00:20:38.370 align:middle line:84% The first is called TANF, Temporary Assistance 00:20:38.370 --> 00:20:39.590 align:middle line:90% to Needy Families. 00:20:39.590 --> 00:20:43.650 align:middle line:90% 00:20:43.650 --> 00:20:48.943 align:middle line:84% TANF is a program that actually started in the Great Depression. 00:20:48.943 --> 00:20:50.610 align:middle line:84% It was originally called Aid to Families 00:20:50.610 --> 00:20:51.830 align:middle line:90% with Dependent Children. 00:20:51.830 --> 00:20:53.850 align:middle line:84% It was actually a program for widows 00:20:53.850 --> 00:20:55.788 align:middle line:84% whose husband had died in World War I 00:20:55.788 --> 00:20:57.330 align:middle line:84% and they couldn't support their kids. 00:20:57.330 --> 00:21:00.410 align:middle line:84% Over time, it became a program, which was really 00:21:00.410 --> 00:21:03.358 align:middle line:84% focused on single mothers, either mothers, women 00:21:03.358 --> 00:21:05.650 align:middle line:84% who are single, sometimes through widowhood, but mostly 00:21:05.650 --> 00:21:10.170 align:middle line:84% through divorce or just out of marriage, childbearing. 00:21:10.170 --> 00:21:13.110 align:middle line:84% And it remains a program which is focused on that population. 00:21:13.110 --> 00:21:15.310 align:middle line:84% Technically, two-parent families can get it as well, 00:21:15.310 --> 00:21:18.270 align:middle line:84% but they're a very small share of the enrollment. 00:21:18.270 --> 00:21:21.190 align:middle line:84% It's a program basically focused on single moms. 00:21:21.190 --> 00:21:23.890 align:middle line:90% How does it work? 00:21:23.890 --> 00:21:27.710 align:middle line:84% Basically, the way it works is, it's a classic cash welfare 00:21:27.710 --> 00:21:28.310 align:middle line:90% scheme. 00:21:28.310 --> 00:21:30.830 align:middle line:90% You get some benefit guarantee. 00:21:30.830 --> 00:21:36.590 align:middle line:84% So the benefits you get are some guaranteed, g, some amount, 00:21:36.590 --> 00:21:41.110 align:middle line:84% and that guarantee is, for example, 00:21:41.110 --> 00:21:44.350 align:middle line:84% $200 a month in Arkansas, all the way up to $1,000 a month 00:21:44.350 --> 00:21:52.390 align:middle line:84% in New Hampshire, so say $200 to $1,000 minus-- 00:21:52.390 --> 00:21:58.150 align:middle line:84% you get a guarantee-- minus a tax rate on your income. 00:21:58.150 --> 00:22:00.950 align:middle line:84% So in other words, if your income 0, you get g. 00:22:00.950 --> 00:22:02.630 align:middle line:90% I should rewrite this. 00:22:02.630 --> 00:22:09.190 align:middle line:84% Let me rewrite this. b equals the max 00:22:09.190 --> 00:22:17.430 align:middle line:84% of 0 comma B minus g minus tau Y, because it can't go negative. 00:22:17.430 --> 00:22:20.330 align:middle line:84% So what you do is, you get a guarantee. 00:22:20.330 --> 00:22:23.490 align:middle line:84% But as you earn income, they reduce how much you get. 00:22:23.490 --> 00:22:26.290 align:middle line:90% So if y is 0, you get g. 00:22:26.290 --> 00:22:30.930 align:middle line:84% But if tau, for example, is 1, as it typically is, 00:22:30.930 --> 00:22:34.210 align:middle line:84% once your income gets to g, you get no more. 00:22:34.210 --> 00:22:36.010 align:middle line:84% So the idea is it's a transfer, but it's 00:22:36.010 --> 00:22:38.490 align:middle line:90% a means-tested transfer. 00:22:38.490 --> 00:22:40.290 align:middle line:84% The higher your means, as measured 00:22:40.290 --> 00:22:44.090 align:middle line:84% by your income, the less money you get from the program. 00:22:44.090 --> 00:22:46.070 align:middle line:90% So it's a means-tested transfer. 00:22:46.070 --> 00:22:51.450 align:middle line:90% 00:22:51.450 --> 00:22:53.490 align:middle line:84% That's one of our major cash transfer programs. 00:22:53.490 --> 00:22:56.210 align:middle line:84% That goes to single moms, so it's means 00:22:56.210 --> 00:22:59.450 align:middle line:90% tested and categorical. 00:22:59.450 --> 00:23:02.250 align:middle line:84% The second big cash transfer program is actually much bigger. 00:23:02.250 --> 00:23:06.850 align:middle line:84% It's Supplemental Security Income, SSI. 00:23:06.850 --> 00:23:11.490 align:middle line:84% Supplemental Security Income is basically 00:23:11.490 --> 00:23:14.710 align:middle line:84% money for those who are disabled but don't qualify for HINDI. 00:23:14.710 --> 00:23:15.210 align:middle line:90% DI. 00:23:15.210 --> 00:23:18.070 align:middle line:84% We talked about DI in chapter 14, 00:23:18.070 --> 00:23:20.950 align:middle line:84% but DI is for people who've worked for a while, 00:23:20.950 --> 00:23:24.110 align:middle line:84% but many, many people never were able to work 00:23:24.110 --> 00:23:26.950 align:middle line:84% or didn't work enough to qualify. 00:23:26.950 --> 00:23:30.650 align:middle line:90% So SSI is categorical. 00:23:30.650 --> 00:23:33.090 align:middle line:84% You have to be disabled, and it's means tested. 00:23:33.090 --> 00:23:36.030 align:middle line:84% You have to be poor and not have enough resources to qualify 00:23:36.030 --> 00:23:38.670 align:middle line:90% for SSDI. 00:23:38.670 --> 00:23:42.830 align:middle line:84% So this is a categorical means-tested program. 00:23:42.830 --> 00:23:45.870 align:middle line:84% Now the debate about welfare in the US 00:23:45.870 --> 00:23:48.430 align:middle line:84% is focused on this program, on TANF. 00:23:48.430 --> 00:23:50.275 align:middle line:90% SSI is 4 times as big as TANF. 00:23:50.275 --> 00:23:51.650 align:middle line:84% It's a much, much bigger program. 00:23:51.650 --> 00:23:53.570 align:middle line:84% This is about a $15 billion program. 00:23:53.570 --> 00:23:57.430 align:middle line:84% This is about a $60 billion program in normal times. 00:23:57.430 --> 00:23:59.795 align:middle line:84% So interestingly, most of our distribution 00:23:59.795 --> 00:24:01.170 align:middle line:84% actually goes to disabled people, 00:24:01.170 --> 00:24:02.545 align:middle line:84% not to the very poor single moms. 00:24:02.545 --> 00:24:05.950 align:middle line:84% But that's been the focus of the welfare debate. 00:24:05.950 --> 00:24:09.950 align:middle line:84% So that's our two big cash programs. 00:24:09.950 --> 00:24:14.360 align:middle line:84% Those are dominated by our in-kind programs. 00:24:14.360 --> 00:24:16.560 align:middle line:84% We spend much more money in the US 00:24:16.560 --> 00:24:20.320 align:middle line:84% on in-kind programs, of which we have a wide variety. 00:24:20.320 --> 00:24:24.080 align:middle line:84% The first, we have SNAP, Supplemental Nutrition 00:24:24.080 --> 00:24:25.480 align:middle line:90% Assistance Program. 00:24:25.480 --> 00:24:28.400 align:middle line:84% When I was growing up, it was called food stamps. 00:24:28.400 --> 00:24:31.360 align:middle line:84% This is a program which, if you're poor-- 00:24:31.360 --> 00:24:34.280 align:middle line:84% think of it as roughly going to people below the poverty line. 00:24:34.280 --> 00:24:36.680 align:middle line:84% If you live below the poverty line, 00:24:36.680 --> 00:24:42.800 align:middle line:84% you get a debit card that you can use to buy food, 00:24:42.800 --> 00:24:43.700 align:middle line:90% that you can use. 00:24:43.700 --> 00:24:49.400 align:middle line:84% So basically what it does is, it says that essentially, we're not 00:24:49.400 --> 00:24:50.800 align:middle line:90% just going to give you cash. 00:24:50.800 --> 00:24:55.360 align:middle line:84% We're going to restrict you to using that card to buy food. 00:24:55.360 --> 00:24:58.680 align:middle line:84% We talked before about why governments may want to do this, 00:24:58.680 --> 00:25:01.120 align:middle line:84% because they're concerned that that money might get spent 00:25:01.120 --> 00:25:03.520 align:middle line:84% on something else, so paternalistically, they 00:25:03.520 --> 00:25:05.840 align:middle line:84% might want to force people to use it on food. 00:25:05.840 --> 00:25:07.860 align:middle line:90% It used to literally be stamps. 00:25:07.860 --> 00:25:10.800 align:middle line:84% Now it's a debit card that you can only use on food. 00:25:10.800 --> 00:25:12.540 align:middle line:90% This is a very big program. 00:25:12.540 --> 00:25:16.040 align:middle line:84% This is a program for costs about $60 billion a year, 00:25:16.040 --> 00:25:19.180 align:middle line:84% comparable to the biggest cash benefit program. 00:25:19.180 --> 00:25:24.300 align:middle line:84% Now to get this, you have to either show 00:25:24.300 --> 00:25:26.820 align:middle line:84% that you're unable to work or that you're looking for work 00:25:26.820 --> 00:25:27.880 align:middle line:90% and have been able to find it. 00:25:27.880 --> 00:25:29.213 align:middle line:90% You can't just qualify for this. 00:25:29.213 --> 00:25:30.420 align:middle line:90% It's somewhat categorical. 00:25:30.420 --> 00:25:32.180 align:middle line:84% It used to be purely means tested 00:25:32.180 --> 00:25:33.620 align:middle line:90% and it was that the poor got it. 00:25:33.620 --> 00:25:35.820 align:middle line:84% Now you have to be poor, but you have 00:25:35.820 --> 00:25:40.140 align:middle line:84% to show that you're registered for work and looking for work. 00:25:40.140 --> 00:25:44.460 align:middle line:84% We'll talk about why they added that condition. 00:25:44.460 --> 00:25:46.020 align:middle line:90% That's SNAP. 00:25:46.020 --> 00:25:49.660 align:middle line:84% Of course, the monster in all this thing is Medicaid. 00:25:49.660 --> 00:25:52.640 align:middle line:84% Medicaid, which is about $700 billion a year, 00:25:52.640 --> 00:25:54.060 align:middle line:90% dominates everything else. 00:25:54.060 --> 00:25:56.940 align:middle line:84% Everything else combined is half the size of Medicaid. 00:25:56.940 --> 00:25:59.840 align:middle line:84% That's the biggest transfer program in the country. 00:25:59.840 --> 00:26:01.200 align:middle line:90% And we've talked about Medicaid. 00:26:01.200 --> 00:26:03.740 align:middle line:84% We don't need to talk more about that. 00:26:03.740 --> 00:26:06.320 align:middle line:90% The third is public housing. 00:26:06.320 --> 00:26:10.600 align:middle line:90% 00:26:10.600 --> 00:26:14.055 align:middle line:84% Now public housing comes in two forms. 00:26:14.055 --> 00:26:15.680 align:middle line:84% The first form is what was colloquially 00:26:15.680 --> 00:26:20.480 align:middle line:84% called the projects, which are big buildings where we literally 00:26:20.480 --> 00:26:24.080 align:middle line:90% give you an apartment. 00:26:24.080 --> 00:26:26.320 align:middle line:84% That's been replaced over time with what's 00:26:26.320 --> 00:26:34.360 align:middle line:84% called Section 8 vouchers, which are basically coupons that you 00:26:34.360 --> 00:26:38.800 align:middle line:90% can take to rent an apartment. 00:26:38.800 --> 00:26:40.520 align:middle line:84% Essentially, how public housing works is, 00:26:40.520 --> 00:26:43.460 align:middle line:90% you have to be low income. 00:26:43.460 --> 00:26:46.825 align:middle line:90% 00:26:46.825 --> 00:26:48.200 align:middle line:84% You typically have to have income 00:26:48.200 --> 00:26:52.222 align:middle line:84% below half of the area income, so this one is area adjusted. 00:26:52.222 --> 00:26:53.180 align:middle line:90% The other ones are not. 00:26:53.180 --> 00:26:54.300 align:middle line:90% This one is area adjusted. 00:26:54.300 --> 00:26:58.720 align:middle line:84% So income has to be below half of the typical area income. 00:26:58.720 --> 00:27:01.680 align:middle line:84% If you're below half the typical area income, 00:27:01.680 --> 00:27:03.920 align:middle line:84% you can either qualify for an apartment 00:27:03.920 --> 00:27:05.760 align:middle line:90% or qualify for a voucher. 00:27:05.760 --> 00:27:07.860 align:middle line:84% The way the apartments work is, you get it, 00:27:07.860 --> 00:27:10.980 align:middle line:84% you pay 30% of your income towards the apartment, 00:27:10.980 --> 00:27:12.820 align:middle line:84% and the government pays the rest. 00:27:12.820 --> 00:27:14.660 align:middle line:90% So it's means tested. 00:27:14.660 --> 00:27:16.180 align:middle line:84% So essentially, the way it works is, 00:27:16.180 --> 00:27:17.880 align:middle line:84% we say we're going to give you an apartment. 00:27:17.880 --> 00:27:19.640 align:middle line:84% You have to contribute 30% of your income. 00:27:19.640 --> 00:27:22.460 align:middle line:84% So if your income is $0, you contribute nothing. 00:27:22.460 --> 00:27:26.630 align:middle line:84% If it becomes very high, you're not qualifying for the program. 00:27:26.630 --> 00:27:28.380 align:middle line:84% And then the government will pay the rest. 00:27:28.380 --> 00:27:29.800 align:middle line:84% Section 8 vouchers are different. 00:27:29.800 --> 00:27:30.967 align:middle line:90% They're kind of interesting. 00:27:30.967 --> 00:27:33.380 align:middle line:84% Section 8 vouchers are-- we give you a piece of paper 00:27:33.380 --> 00:27:34.920 align:middle line:90% you can take to your landlord. 00:27:34.920 --> 00:27:36.780 align:middle line:84% It says to the landlord, the government 00:27:36.780 --> 00:27:39.460 align:middle line:84% will pay any amount of the rent above 30% 00:27:39.460 --> 00:27:40.827 align:middle line:90% of this person's income. 00:27:40.827 --> 00:27:42.160 align:middle line:90% So you have a Section 8 voucher. 00:27:42.160 --> 00:27:43.200 align:middle line:90% You go to landlord. 00:27:43.200 --> 00:27:45.860 align:middle line:84% The landlord collects 30% of your income as rent from you 00:27:45.860 --> 00:27:48.820 align:middle line:84% and the rest from the government. 00:27:48.820 --> 00:27:51.320 align:middle line:84% This is increasingly what we're doing with public housing. 00:27:51.320 --> 00:27:54.980 align:middle line:84% We're increasingly moving away from projects, towards that. 00:27:54.980 --> 00:28:00.260 align:middle line:84% And this is about a $30 billion a year program. 00:28:00.260 --> 00:28:02.440 align:middle line:84% Then, the last category, which is a big one, 00:28:02.440 --> 00:28:10.670 align:middle line:84% is a whole set of nutritional programs. 00:28:10.670 --> 00:28:15.150 align:middle line:84% This includes things like free or reduced price school meals, 00:28:15.150 --> 00:28:18.310 align:middle line:84% school breakfast and lunch, which 00:28:18.310 --> 00:28:22.230 align:middle line:84% is a major source of redistribution in our country. 00:28:22.230 --> 00:28:26.990 align:middle line:84% We spend about $20 billion a year on these. 00:28:26.990 --> 00:28:30.430 align:middle line:84% In a typical year, about 22 million people 00:28:30.430 --> 00:28:31.970 align:middle line:84% get free or reduced price lunches. 00:28:31.970 --> 00:28:35.150 align:middle line:84% About 13 million get free or reduced price breakfasts. 00:28:35.150 --> 00:28:38.270 align:middle line:84% So it's a major program they spend about $20 billion a year 00:28:38.270 --> 00:28:38.810 align:middle line:90% on. 00:28:38.810 --> 00:28:40.950 align:middle line:84% There's also something called the WIC program, 00:28:40.950 --> 00:28:43.010 align:middle line:84% which program for Women, Infants and Children, 00:28:43.010 --> 00:28:44.453 align:middle line:84% which gives them the actual food. 00:28:44.453 --> 00:28:45.370 align:middle line:90% It's like food stamps. 00:28:45.370 --> 00:28:46.787 align:middle line:84% But instead of giving you a debit, 00:28:46.787 --> 00:28:48.570 align:middle line:84% they literally give out milk, cheese, 00:28:48.570 --> 00:28:52.870 align:middle line:84% and other healthy products for women, infants, and children. 00:28:52.870 --> 00:28:56.510 align:middle line:84% The bottom line is, actually, the vast majority 00:28:56.510 --> 00:28:59.790 align:middle line:84% of redistribution we do is in kind. 00:28:59.790 --> 00:29:03.230 align:middle line:84% Even the very biggest cash program 00:29:03.230 --> 00:29:06.370 align:middle line:84% is smaller than a lot of these in-kind programs. 00:29:06.370 --> 00:29:10.510 align:middle line:84% So the interesting question will come to is why? 00:29:10.510 --> 00:29:12.770 align:middle line:84% We'll come to that later in this lecture. 00:29:12.770 --> 00:29:15.590 align:middle line:84% So that's the way it works in a nutshell. 00:29:15.590 --> 00:29:16.390 align:middle line:90% Yeah? 00:29:16.390 --> 00:29:19.275 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:29:19.275 --> 00:29:20.150 align:middle line:90% PROFESSOR: Excellent. 00:29:20.150 --> 00:29:26.210 align:middle line:84% The voucher is set equal to the maximum of the rent or I think 00:29:26.210 --> 00:29:29.610 align:middle line:84% it's the 40th percentile of the housing cost distribution 00:29:29.610 --> 00:29:30.610 align:middle line:90% in the city. 00:29:30.610 --> 00:29:32.670 align:middle line:84% They cap it, so you can't charge anything. 00:29:32.670 --> 00:29:33.670 align:middle line:90% They give you a voucher. 00:29:33.670 --> 00:29:36.130 align:middle line:84% You have to find an apartment, which is up to the 40th 00:29:36.130 --> 00:29:39.450 align:middle line:84% percentile, and then you get a voucher for that. 00:29:39.450 --> 00:29:40.783 align:middle line:90% Yeah? 00:29:40.783 --> 00:29:42.450 align:middle line:84% AUDIENCE: Affordable housing [INAUDIBLE] 00:29:42.450 --> 00:29:45.410 align:middle line:84% that has, for example, states like Massachusetts 00:29:45.410 --> 00:29:51.830 align:middle line:84% would be the [INAUDIBLE] It's also an example of it, right? 00:29:51.830 --> 00:29:52.830 align:middle line:90% PROFESSOR: Say it again. 00:29:52.830 --> 00:29:53.550 align:middle line:90% It's also what? 00:29:53.550 --> 00:29:55.050 align:middle line:84% AUDIENCE: An example of [INAUDIBLE]? 00:29:55.050 --> 00:29:55.410 align:middle line:90% PROFESSOR: Yeah. 00:29:55.410 --> 00:29:57.493 align:middle line:84% So basically, how do they allocate these vouchers? 00:29:57.493 --> 00:29:59.050 align:middle line:90% They do it by lottery. 00:29:59.050 --> 00:30:00.170 align:middle line:90% Yeah. 00:30:00.170 --> 00:30:01.765 align:middle line:84% They basically have a waiting list. 00:30:01.765 --> 00:30:03.390 align:middle line:84% You lottery to get onto a waiting list, 00:30:03.390 --> 00:30:04.190 align:middle line:84% and they pull you off the waiting 00:30:04.190 --> 00:30:05.650 align:middle line:90% list as units become available. 00:30:05.650 --> 00:30:09.910 align:middle line:90% 00:30:09.910 --> 00:30:13.710 align:middle line:84% So with that as background, let's talk 00:30:13.710 --> 00:30:16.310 align:middle line:84% about how we think about the optimal determination 00:30:16.310 --> 00:30:17.870 align:middle line:90% of transfer programs. 00:30:17.870 --> 00:30:20.430 align:middle line:84% Let's say we've decided we want to transfer money. 00:30:20.430 --> 00:30:23.790 align:middle line:84% How do we think about the efficiency consequences 00:30:23.790 --> 00:30:24.930 align:middle line:90% of a transfer program? 00:30:24.930 --> 00:30:29.190 align:middle line:84% And here we come to the core of thinking 00:30:29.190 --> 00:30:31.830 align:middle line:84% about equity and redistribution economics, which 00:30:31.830 --> 00:30:34.868 align:middle line:84% is the efficiency-equity trade-off. 00:30:34.868 --> 00:30:37.410 align:middle line:84% And so we're going to embody the efficiency-equity trade-off. 00:30:37.410 --> 00:30:38.790 align:middle line:84% And the best way to think about that-- 00:30:38.790 --> 00:30:40.110 align:middle line:84% one of my favorite concepts in economics-- 00:30:40.110 --> 00:30:41.750 align:middle line:84% I talked about this in the 14.01-- 00:30:41.750 --> 00:30:44.070 align:middle line:84% is the concept of Okun's leaky bucket. 00:30:44.070 --> 00:30:46.050 align:middle line:84% Arthur Okun was a famous economist. 00:30:46.050 --> 00:30:48.590 align:middle line:84% Arthur Okun said, think about transferring money 00:30:48.590 --> 00:30:50.090 align:middle line:84% for a rich person or a poor person, 00:30:50.090 --> 00:30:52.070 align:middle line:84% as if a government bureaucrat goes and collects the money 00:30:52.070 --> 00:30:54.070 align:middle line:84% from the rich person, carries it in a bucket, 00:30:54.070 --> 00:30:56.412 align:middle line:84% and dumps it out in front of the poor person. 00:30:56.412 --> 00:30:57.870 align:middle line:84% So the reason that's useful is then 00:30:57.870 --> 00:31:01.390 align:middle line:84% you could say, well, imagine that for every dollar you 00:31:01.390 --> 00:31:04.130 align:middle line:84% he took from the rich person, he brought $1 to the poor person. 00:31:04.130 --> 00:31:05.230 align:middle line:90% Would you agree with that? 00:31:05.230 --> 00:31:06.150 align:middle line:90% And most people would say, yeah. 00:31:06.150 --> 00:31:07.120 align:middle line:84% If you can get $1 from a rich person 00:31:07.120 --> 00:31:09.537 align:middle line:84% and give it to a poor person, that's probably worth doing. 00:31:09.537 --> 00:31:12.032 align:middle line:84% He said, but what if there was a leak in the bucket? 00:31:12.032 --> 00:31:13.990 align:middle line:84% What if from every dollar from the rich person, 00:31:13.990 --> 00:31:16.530 align:middle line:84% only $0.80 got to the poor person because $0.20 leaked 00:31:16.530 --> 00:31:17.490 align:middle line:90% along the way? 00:31:17.490 --> 00:31:19.090 align:middle line:90% I might say, well, that's OK. 00:31:19.090 --> 00:31:23.042 align:middle line:90% $0.70, $0.50. $0.10, 0? 00:31:23.042 --> 00:31:24.750 align:middle line:84% Some people might say it's OK, even at 0. 00:31:24.750 --> 00:31:27.190 align:middle line:84% If you think relative income inequality is what matters, 00:31:27.190 --> 00:31:29.770 align:middle line:84% it's OK to just destroy wealth at the top, which 00:31:29.770 --> 00:31:32.930 align:middle line:84% is why people are queasy about endorsing relative income 00:31:32.930 --> 00:31:34.890 align:middle line:90% inequality as a concept. 00:31:34.890 --> 00:31:39.370 align:middle line:84% But basically, what Okun said is your taste for redistribution 00:31:39.370 --> 00:31:41.130 align:middle line:84% can be summarized at how much leak 00:31:41.130 --> 00:31:43.237 align:middle line:84% you're willing to put up with in the bucket. 00:31:43.237 --> 00:31:45.070 align:middle line:84% The more leak you're willing to put up with, 00:31:45.070 --> 00:31:46.730 align:middle line:84% the more you care about the poor, relative to the rich, 00:31:46.730 --> 00:31:49.010 align:middle line:84% getting money from the rich to the poor. 00:31:49.010 --> 00:31:51.730 align:middle line:84% Some people would say, don't even do it with no leak. 00:31:51.730 --> 00:31:53.970 align:middle line:84% Some people would say, do it with a 100% leak, 00:31:53.970 --> 00:31:56.330 align:middle line:90% and others will land in between. 00:31:56.330 --> 00:31:58.630 align:middle line:84% That's just a great concept for thinking about it. 00:31:58.630 --> 00:32:00.830 align:middle line:84% And what it does is it highlights what 00:32:00.830 --> 00:32:02.570 align:middle line:90% is the problem of distribution. 00:32:02.570 --> 00:32:04.630 align:middle line:90% It's the leak in the bucket. 00:32:04.630 --> 00:32:07.080 align:middle line:84% And where does the leak in the bucket come from? 00:32:07.080 --> 00:32:09.330 align:middle line:84% Well, the leak in the bucket comes from three sources. 00:32:09.330 --> 00:32:10.510 align:middle line:84% The first and smallest source is that you've 00:32:10.510 --> 00:32:12.407 align:middle line:84% got to hire someone to carry the bucket. 00:32:12.407 --> 00:32:13.990 align:middle line:84% If you're going to distribute, there's 00:32:13.990 --> 00:32:16.570 align:middle line:84% got to be a redistributive administrative mechanism, 00:32:16.570 --> 00:32:17.530 align:middle line:90% and that costs money. 00:32:17.530 --> 00:32:22.115 align:middle line:84% It turns out, that's not a big deal, but it's something. 00:32:22.115 --> 00:32:23.990 align:middle line:84% The second source of leak in the bucket we'll 00:32:23.990 --> 00:32:26.590 align:middle line:84% talk about starting in a couple of lectures 00:32:26.590 --> 00:32:28.970 align:middle line:84% from now, which is to raise the money. 00:32:28.970 --> 00:32:30.170 align:middle line:90% You have to tax people. 00:32:30.170 --> 00:32:31.870 align:middle line:90% You have to tax the rich. 00:32:31.870 --> 00:32:33.630 align:middle line:84% And when you tax the rich, that can 00:32:33.630 --> 00:32:36.350 align:middle line:84% have distortionary consequences for efficiency, 00:32:36.350 --> 00:32:39.350 align:middle line:84% and we'll come to that in a couple of lectures. 00:32:39.350 --> 00:32:41.110 align:middle line:84% What I'll focus on now is the third source 00:32:41.110 --> 00:32:44.030 align:middle line:84% of leak in the bucket, which is that if you transfer 00:32:44.030 --> 00:32:47.430 align:middle line:84% to the poor, that could cause them to reduce 00:32:47.430 --> 00:32:49.910 align:middle line:90% their economic activity. 00:32:49.910 --> 00:32:54.310 align:middle line:84% So to see this, let's talk about a pure means-tested transfer 00:32:54.310 --> 00:32:59.980 align:middle line:84% program, a pure means-tested transfer transform like TANF. 00:32:59.980 --> 00:33:02.615 align:middle line:84% let's Talk about a program like TANF, 00:33:02.615 --> 00:33:05.240 align:middle line:84% where the government's going to come to you and say-- and let's 00:33:05.240 --> 00:33:07.840 align:middle line:90% set tau equal to 1. 00:33:07.840 --> 00:33:10.420 align:middle line:84% So the government's coming to you and saying, 00:33:10.420 --> 00:33:12.320 align:middle line:90% I'm going to give you money. 00:33:12.320 --> 00:33:16.120 align:middle line:84% As long as your income is below g, you're going to get money. 00:33:16.120 --> 00:33:18.360 align:middle line:84% And that money is going to depend on your income. 00:33:18.360 --> 00:33:19.943 align:middle line:84% For every dollar of income, I give you 00:33:19.943 --> 00:33:21.060 align:middle line:90% one less dollar in grant. 00:33:21.060 --> 00:33:25.400 align:middle line:90% 00:33:25.400 --> 00:33:27.880 align:middle line:84% First of all, that is actually an easy to understand 00:33:27.880 --> 00:33:29.620 align:middle line:84% and relatively attractive system. 00:33:29.620 --> 00:33:33.440 align:middle line:84% It turns out, under this system, if I took everyone in the US 00:33:33.440 --> 00:33:36.440 align:middle line:84% who lives below poverty and implemented 00:33:36.440 --> 00:33:39.520 align:middle line:84% this system at their income today, 00:33:39.520 --> 00:33:43.160 align:middle line:84% I could end poverty in the US for less than $200 billion. 00:33:43.160 --> 00:33:45.440 align:middle line:84% We could have no more poor people in the US, 00:33:45.440 --> 00:33:48.440 align:middle line:84% according to the poverty line, for $180 billion. 00:33:48.440 --> 00:33:50.100 align:middle line:84% That's pretty good, I mean, every year, 00:33:50.100 --> 00:33:53.180 align:middle line:84% but we spent more than that on other things. 00:33:53.180 --> 00:33:55.080 align:middle line:90% That's not a crazy number. 00:33:55.080 --> 00:33:58.180 align:middle line:84% The problem with that number is it 00:33:58.180 --> 00:34:00.540 align:middle line:84% misses the moral hazard consequence of actually doing 00:34:00.540 --> 00:34:02.020 align:middle line:90% that transfer. 00:34:02.020 --> 00:34:06.700 align:middle line:84% And that's illustrated well in figure 17.4. 00:34:06.700 --> 00:34:09.860 align:middle line:84% Here we have an individual's decision on how hard to work. 00:34:09.860 --> 00:34:14.219 align:middle line:84% They're doing the standard leisure consumption trade-off. 00:34:14.219 --> 00:34:18.460 align:middle line:84% On the x-axis, you have hours of leisure per year. 00:34:18.460 --> 00:34:21.139 align:middle line:84% On the y-axis, you have dollars of consumption. 00:34:21.139 --> 00:34:24.940 align:middle line:84% And imagine they earn 15 bucks an hour. 00:34:24.940 --> 00:34:28.219 align:middle line:84% And imagine there's 2,000 hours a year to work. 00:34:28.219 --> 00:34:31.957 align:middle line:84% So their budget constraint, which runs from A to C, 00:34:31.957 --> 00:34:34.540 align:middle line:84% says that they could either have $30,000 of consumption and no 00:34:34.540 --> 00:34:38.260 align:middle line:84% leisure or 2,000 hours of leisure and no consumption 00:34:38.260 --> 00:34:41.060 align:middle line:84% or some combination in between, with a marginal rate 00:34:41.060 --> 00:34:43.980 align:middle line:84% of transformation of 15 bucks an hour. 00:34:43.980 --> 00:34:47.739 align:middle line:84% Now imagine we come into that world 00:34:47.739 --> 00:34:50.011 align:middle line:84% and we put in the following system. 00:34:50.011 --> 00:34:51.719 align:middle line:84% We say we're going to put in this system, 00:34:51.719 --> 00:34:54.940 align:middle line:84% and we're going to set g equal to the poverty line, 00:34:54.940 --> 00:34:56.620 align:middle line:90% which is $12,760. 00:34:56.620 --> 00:34:58.480 align:middle line:84% Let's think about individuals here. 00:34:58.480 --> 00:35:03.560 align:middle line:84% So we're going to implement this system with g equal to 12,760 00:35:03.560 --> 00:35:06.600 align:middle line:90% and tau equal to 1. 00:35:06.600 --> 00:35:09.080 align:middle line:84% What would that do to the budget constraint? 00:35:09.080 --> 00:35:13.140 align:middle line:84% What it would do is, it would add a flat segment running 00:35:13.140 --> 00:35:16.040 align:middle line:84% so the new budget from A to B would be unchanged. 00:35:16.040 --> 00:35:19.040 align:middle line:84% But at point B, you'd have a flat segment over to D 00:35:19.040 --> 00:35:22.040 align:middle line:90% and then down to C. Why? 00:35:22.040 --> 00:35:27.700 align:middle line:84% Because once your income is below $12,760, or alternatively, 00:35:27.700 --> 00:35:31.960 align:middle line:84% once you're taking more than 1,149 hours of leisure, 00:35:31.960 --> 00:35:34.620 align:middle line:84% each additional dollar of earnings does nothing for you. 00:35:34.620 --> 00:35:35.120 align:middle line:90% Why? 00:35:35.120 --> 00:35:37.800 align:middle line:84% Because the government just taxes it away. 00:35:37.800 --> 00:35:42.920 align:middle line:84% So it doesn't matter if I work 1,600 hours or if I take 1,600 00:35:42.920 --> 00:35:46.500 align:middle line:84% hours of leisure, or 1,149 hours of leisure. 00:35:46.500 --> 00:35:47.720 align:middle line:90% I get the same thing. 00:35:47.720 --> 00:35:50.240 align:middle line:90% I get $12,760. 00:35:50.240 --> 00:35:54.420 align:middle line:84% So the budget constraint becomes flat at that point. 00:35:54.420 --> 00:35:56.580 align:middle line:90% Questions about that? 00:35:56.580 --> 00:35:58.780 align:middle line:84% Now think about what that does to labor supply. 00:35:58.780 --> 00:36:01.740 align:middle line:90% Let's start with person x. 00:36:01.740 --> 00:36:07.540 align:middle line:84% Person x was making only $6,000 and working only 400 hours 00:36:07.540 --> 00:36:11.420 align:middle line:90% a year, 1,600 hours of leisure. 00:36:11.420 --> 00:36:18.700 align:middle line:84% Person x unambiguously will move to point D. That's 00:36:18.700 --> 00:36:20.940 align:middle line:90% a dominant strategy for them. 00:36:20.940 --> 00:36:24.380 align:middle line:84% They get more leisure and more consumption. 00:36:24.380 --> 00:36:29.020 align:middle line:84% So unambiguously, they move to point D. 00:36:29.020 --> 00:36:34.660 align:middle line:84% On the other hand, now let's take someone like person z. 00:36:34.660 --> 00:36:38.900 align:middle line:84% Person z will not be affected by this welfare program. 00:36:38.900 --> 00:36:39.820 align:middle line:90% Why? 00:36:39.820 --> 00:36:43.700 align:middle line:84% Because they like consumption so much that a program which leaves 00:36:43.700 --> 00:36:46.710 align:middle line:84% them with $12,760 of income is irrelevant. 00:36:46.710 --> 00:36:48.460 align:middle line:84% They're not willing to live at that level. 00:36:48.460 --> 00:36:51.190 align:middle line:84% They'd rather just go ahead and work and have their $22,000 00:36:51.190 --> 00:36:53.510 align:middle line:90% of consumption. 00:36:53.510 --> 00:36:54.770 align:middle line:90% What about person y? 00:36:54.770 --> 00:36:58.230 align:middle line:84% Well, person y is purposely and painstakingly drawn, 00:36:58.230 --> 00:37:01.030 align:middle line:84% so that their indifference curve-- 00:37:01.030 --> 00:37:02.720 align:middle line:84% it's not really a very good curve. 00:37:02.720 --> 00:37:03.970 align:middle line:90% It's hard to draw it this way. 00:37:03.970 --> 00:37:07.910 align:middle line:84% But basically, the point is, it's tangent at y. 00:37:07.910 --> 00:37:14.550 align:middle line:84% They originally took 1,067 hours of leisure and consumed 14,000. 00:37:14.550 --> 00:37:18.710 align:middle line:84% But their indifference curve cuts below the horizontal line 00:37:18.710 --> 00:37:23.310 align:middle line:84% from B to D. As a result, moving to point D gives them 00:37:23.310 --> 00:37:26.423 align:middle line:90% a higher level of utility. 00:37:26.423 --> 00:37:28.590 align:middle line:84% You can see that since their indifference curve cuts 00:37:28.590 --> 00:37:32.590 align:middle line:84% from above that flat line from B to D, that at point D, 00:37:32.590 --> 00:37:35.030 align:middle line:90% they have a higher utility. 00:37:35.030 --> 00:37:37.150 align:middle line:90% So they will quit. 00:37:37.150 --> 00:37:39.750 align:middle line:84% They will be willing to consume somewhat less 00:37:39.750 --> 00:37:42.510 align:middle line:84% to get a lot more leisure, just like the Peltzman diagram 00:37:42.510 --> 00:37:44.470 align:middle line:90% for education crowd out. 00:37:44.470 --> 00:37:50.470 align:middle line:84% It's a similar crowded argument here, more than 100% crowd out. 00:37:50.470 --> 00:37:55.050 align:middle line:84% You'll literally take away all their labor supply. 00:37:55.050 --> 00:37:57.543 align:middle line:84% And not just you're not just take away the labor 00:37:57.543 --> 00:37:58.710 align:middle line:90% supply you're taking from x. 00:37:58.710 --> 00:38:00.710 align:middle line:84% You'll take even more than that, because they'll 00:38:00.710 --> 00:38:05.210 align:middle line:84% drop down their consumption a bit to qualify for this program. 00:38:05.210 --> 00:38:08.460 align:middle line:90% Questions about that? 00:38:08.460 --> 00:38:10.210 align:middle line:84% Think about what this does to our estimate 00:38:10.210 --> 00:38:12.170 align:middle line:90% of the cost of the program. 00:38:12.170 --> 00:38:15.650 align:middle line:84% If we just consider people like x 00:38:15.650 --> 00:38:17.835 align:middle line:84% and realize that people below the poverty line 00:38:17.835 --> 00:38:19.960 align:middle line:84% aren't going to-- we aren't going to give them this 00:38:19.960 --> 00:38:21.418 align:middle line:84% and we're just going to give them g 00:38:21.418 --> 00:38:25.170 align:middle line:84% because they're all going to quit, the cost doubles. 00:38:25.170 --> 00:38:29.230 align:middle line:90% If we consider people like y-- 00:38:29.230 --> 00:38:31.450 align:middle line:84% let's say people living 25% above the poverty line 00:38:31.450 --> 00:38:35.050 align:middle line:84% react that way, that adds another 50% to the cost. 00:38:35.050 --> 00:38:37.450 align:middle line:84% So now you're talking about a really expensive program 00:38:37.450 --> 00:38:39.730 align:middle line:84% through these moral hazard effects. 00:38:39.730 --> 00:38:41.770 align:middle line:84% The point is that these moral hazard effects 00:38:41.770 --> 00:38:45.770 align:middle line:84% make redistribution a lot more costly than you 00:38:45.770 --> 00:38:48.750 align:middle line:90% would think in their absence. 00:38:48.750 --> 00:38:53.630 align:middle line:84% And that's the standard moral hazard problem 00:38:53.630 --> 00:38:56.270 align:middle line:84% or the leak in Okun's bucket that 00:38:56.270 --> 00:39:00.350 align:middle line:84% comes from moral hazard of poor people 00:39:00.350 --> 00:39:06.850 align:middle line:84% lowering their resources to qualify for a program. 00:39:06.850 --> 00:39:11.190 align:middle line:84% Now it seems there's a simple answer to this problem, which 00:39:11.190 --> 00:39:12.730 align:middle line:90% is, well, the problem is this. 00:39:12.730 --> 00:39:16.630 align:middle line:84% Why make money with 100% tax rate? 00:39:16.630 --> 00:39:20.090 align:middle line:84% We should lower the tax rate, so it's an incentive to work. 00:39:20.090 --> 00:39:22.670 align:middle line:84% We should redistribute, but let's have an incentive to work. 00:39:22.670 --> 00:39:27.350 align:middle line:84% Indeed, what if we change this 1 to 0.5? 00:39:27.350 --> 00:39:32.790 align:middle line:84% Well, then we would get a figure like Figure 17.5. 00:39:32.790 --> 00:39:41.670 align:middle line:84% Figure 17.5 says that for consumption above $25,520 00:39:41.670 --> 00:39:44.730 align:middle line:84% or leisure of less than 299 hours a year, 00:39:44.730 --> 00:39:47.130 align:middle line:90% the budget rate hasn't changed. 00:39:47.130 --> 00:39:49.570 align:middle line:84% Starting at that point, the budget rate 00:39:49.570 --> 00:39:52.850 align:middle line:84% now has a slope that's half as steep. 00:39:52.850 --> 00:39:53.710 align:middle line:90% Why half as steep? 00:39:53.710 --> 00:39:56.230 align:middle line:84% Because essentially, there's a 50% tax on your set wage. 00:39:56.230 --> 00:39:57.772 align:middle line:84% Essentially, for every hour you work, 00:39:57.772 --> 00:39:59.630 align:middle line:84% you take home $7.50 instead of $15. 00:39:59.630 --> 00:40:00.130 align:middle line:90% Why? 00:40:00.130 --> 00:40:03.010 align:middle line:84% Because you give up that government transfer. 00:40:03.010 --> 00:40:05.370 align:middle line:90% It's an implicit tax of 50%. 00:40:05.370 --> 00:40:08.650 align:middle line:84% So the line, which intersects at B2, 00:40:08.650 --> 00:40:15.010 align:middle line:84% the a slope that's 50% as steep, $7.50, all the way to point D. 00:40:15.010 --> 00:40:18.770 align:middle line:84% What does that do to people's decisions? 00:40:18.770 --> 00:40:23.330 align:middle line:84% Well, for person x, they work more, 00:40:23.330 --> 00:40:26.890 align:middle line:84% or they might work more because now, 00:40:26.890 --> 00:40:29.090 align:middle line:90% you've lowered their tax rate. 00:40:29.090 --> 00:40:32.610 align:middle line:84% For person y, they might work more as well 00:40:32.610 --> 00:40:34.028 align:middle line:90% because you lower the tax rate. 00:40:34.028 --> 00:40:35.570 align:middle line:84% So if you stop there, you'd say, well 00:40:35.570 --> 00:40:37.770 align:middle line:84% this is unambiguously a better system. 00:40:37.770 --> 00:40:40.730 align:middle line:84% The problem is, now z is in the game. 00:40:40.730 --> 00:40:43.650 align:middle line:84% Now z, who wasn't on welfare before, suddenly says, 00:40:43.650 --> 00:40:45.410 align:middle line:84% well, I want to be in welfare now, too. 00:40:45.410 --> 00:40:45.670 align:middle line:90% Why? 00:40:45.670 --> 00:40:47.990 align:middle line:84% Because I don't have to give up that much consumption and I 00:40:47.990 --> 00:40:49.115 align:middle line:90% can get a lot more leisure. 00:40:49.115 --> 00:40:50.870 align:middle line:90% I can move from Z1 to Z2. 00:40:50.870 --> 00:40:52.745 align:middle line:84% I don't have to give up that much consumption 00:40:52.745 --> 00:40:54.790 align:middle line:90% and I get a lot more leisure. 00:40:54.790 --> 00:40:58.630 align:middle line:84% So the problem is, yes, you save money in x and y, 00:40:58.630 --> 00:41:00.870 align:middle line:90% but you spent money on z. 00:41:00.870 --> 00:41:03.630 align:middle line:84% You've brought more people into the program. 00:41:03.630 --> 00:41:05.710 align:middle line:84% And whether, on net, you save money or not 00:41:05.710 --> 00:41:09.710 align:middle line:84% depends on the relative size of these reactions. 00:41:09.710 --> 00:41:14.070 align:middle line:84% This is what we call the iron law of welfare policy, which 00:41:14.070 --> 00:41:17.430 align:middle line:90% is that fundamentally-- 00:41:17.430 --> 00:41:18.210 align:middle line:90% the iron triangle. 00:41:18.210 --> 00:41:19.190 align:middle line:90% I'm sorry. 00:41:19.190 --> 00:41:21.390 align:middle line:84% Fundamentally, we want to try the iron 00:41:21.390 --> 00:41:23.070 align:middle line:90% triangle of welfare policy. 00:41:23.070 --> 00:41:28.310 align:middle line:84% We fundamentally want to try to accomplish three things. 00:41:28.310 --> 00:41:31.698 align:middle line:84% We want to-- if I reform my welfare system, 00:41:31.698 --> 00:41:33.490 align:middle line:84% my existing welfare system, if I change it, 00:41:33.490 --> 00:41:36.230 align:middle line:84% I'm trying to accomplish three things simultaneously. 00:41:36.230 --> 00:41:41.600 align:middle line:84% I want to encourage work that has reduced the moral hazard 00:41:41.600 --> 00:41:44.000 align:middle line:90% costs. 00:41:44.000 --> 00:41:45.800 align:middle line:90% I want to encourage work. 00:41:45.800 --> 00:41:54.360 align:middle line:84% I want to redistribute income, and I 00:41:54.360 --> 00:41:56.180 align:middle line:90% want to lower government costs. 00:41:56.180 --> 00:42:00.160 align:middle line:90% 00:42:00.160 --> 00:42:02.720 align:middle line:84% So I have three goals I'm trying to achieve, 00:42:02.720 --> 00:42:07.440 align:middle line:84% but I only have two instruments, the tax rate and g. 00:42:07.440 --> 00:42:09.440 align:middle line:84% So I'm trying to accomplish three goals with two 00:42:09.440 --> 00:42:11.975 align:middle line:90% instruments, and I can't do it. 00:42:11.975 --> 00:42:14.100 align:middle line:84% In other words, if I gave up on one of these goals, 00:42:14.100 --> 00:42:15.760 align:middle line:90% I could do it. 00:42:15.760 --> 00:42:18.660 align:middle line:84% If I, for example, said I don't care about redistributing, 00:42:18.660 --> 00:42:21.077 align:middle line:84% well, I could encourage work and save the government money 00:42:21.077 --> 00:42:22.840 align:middle line:90% by just having a lower g. 00:42:22.840 --> 00:42:25.880 align:middle line:84% If I had a lower g, I would encourage work 00:42:25.880 --> 00:42:29.960 align:middle line:84% and save the government money, but I redistribute less income. 00:42:29.960 --> 00:42:32.760 align:middle line:84% And you could see how you could use g and tau to solve 00:42:32.760 --> 00:42:35.600 align:middle line:84% any two of these three problems, but you can't solve all three 00:42:35.600 --> 00:42:37.640 align:middle line:90% simultaneously. 00:42:37.640 --> 00:42:41.720 align:middle line:84% And that is the fundamental challenge of welfare policy. 00:42:41.720 --> 00:42:45.260 align:middle line:84% This iron triangle problem was a standard cash welfare program 00:42:45.260 --> 00:42:46.160 align:middle line:90% you cannot solve. 00:42:46.160 --> 00:42:49.500 align:middle line:90% 00:42:49.500 --> 00:42:52.940 align:middle line:90% Questions about that? 00:42:52.940 --> 00:42:57.103 align:middle line:84% So that is what motivates the really interesting work. 00:42:57.103 --> 00:42:58.520 align:middle line:84% There's a lot of interesting work, 00:42:58.520 --> 00:43:01.300 align:middle line:84% of course, in thinking about how bad this moral hazard 00:43:01.300 --> 00:43:02.500 align:middle line:90% problem is. 00:43:02.500 --> 00:43:04.868 align:middle line:84% But other than that, it's a pretty simple concept. 00:43:04.868 --> 00:43:07.160 align:middle line:84% The more interesting thing is, how do we get around it? 00:43:07.160 --> 00:43:07.840 align:middle line:90% Yeah, Enoch? 00:43:07.840 --> 00:43:11.700 align:middle line:84% AUDIENCE: [INAUDIBLE] the thoughts around UBI. 00:43:11.700 --> 00:43:15.660 align:middle line:84% PROFESSOR: I'll come back to UBI the next lecture. 00:43:15.660 --> 00:43:19.140 align:middle line:90% So how do we get around this? 00:43:19.140 --> 00:43:28.020 align:middle line:84% Well, one way is to move towards categorical welfare, 00:43:28.020 --> 00:43:30.860 align:middle line:84% and that's by the following thought experiment. 00:43:30.860 --> 00:43:33.980 align:middle line:84% At the end of the day, there's two reasons why my income might 00:43:33.980 --> 00:43:35.740 align:middle line:90% differ from Valerie's. 00:43:35.740 --> 00:43:38.120 align:middle line:84% One is because I'm more skilled, and two 00:43:38.120 --> 00:43:40.860 align:middle line:90% is because I work harder. 00:43:40.860 --> 00:43:43.440 align:middle line:84% If we think skill is something you're born with, 00:43:43.440 --> 00:43:46.200 align:middle line:84% we want to redistribute from the high skill to low skills. 00:43:46.200 --> 00:43:47.497 align:middle line:90% It's basically luck. 00:43:47.497 --> 00:43:50.080 align:middle line:84% If you're born lucky that you're a high-skilled person, if I'm 00:43:50.080 --> 00:43:51.400 align:middle line:84% born lucky, a high skill person, poor Valerie 00:43:51.400 --> 00:43:53.500 align:middle line:84% is a low-skilled person, she's born unlucky, 00:43:53.500 --> 00:43:56.200 align:middle line:84% we'd like to redistribute from me to her. 00:43:56.200 --> 00:43:59.300 align:middle line:84% But if it's that I'm working hard and she's lazy, 00:43:59.300 --> 00:44:01.360 align:middle line:84% we maybe don't want to shoot from me to her. 00:44:01.360 --> 00:44:03.000 align:middle line:90% That's moral hazard. 00:44:03.000 --> 00:44:05.480 align:middle line:84% So since that's the problem we have here, 00:44:05.480 --> 00:44:08.480 align:middle line:84% that we're trying to distribute on a concept income which 00:44:08.480 --> 00:44:11.460 align:middle line:84% embodies two different things, the thing we care about, 00:44:11.460 --> 00:44:14.480 align:middle line:84% which is who has the ability and who doesn't and the thing 00:44:14.480 --> 00:44:18.660 align:middle line:84% we don't want to distort, which is how hard people work-- 00:44:18.660 --> 00:44:20.160 align:middle line:84% Well, imagine a world where everyone 00:44:20.160 --> 00:44:22.960 align:middle line:84% was born with an indelible tattoo on their forehead that 00:44:22.960 --> 00:44:27.040 align:middle line:84% said, high ability, low ability, and there was no way 00:44:27.040 --> 00:44:30.240 align:middle line:84% to scrub it off, no way to change it, well, then we could 00:44:30.240 --> 00:44:31.843 align:middle line:90% solve the moral hazard problem. 00:44:31.843 --> 00:44:34.260 align:middle line:84% We'd say, you've got a low ability thing on your forehead. 00:44:34.260 --> 00:44:35.060 align:middle line:90% Here's some money. 00:44:35.060 --> 00:44:36.060 align:middle line:90% You've got high ability. 00:44:36.060 --> 00:44:38.140 align:middle line:90% You get no money. 00:44:38.140 --> 00:44:43.460 align:middle line:84% That solves the problem because if you redistribute 00:44:43.460 --> 00:44:45.860 align:middle line:84% on something people can't change, 00:44:45.860 --> 00:44:48.180 align:middle line:90% then there's no moral hazard. 00:44:48.180 --> 00:44:50.980 align:middle line:84% Moral hazard arises from changing your behavior 00:44:50.980 --> 00:44:53.380 align:middle line:84% to take advantage of a government program. 00:44:53.380 --> 00:44:56.140 align:middle line:84% But if I redistribute something you can't change, 00:44:56.140 --> 00:44:58.435 align:middle line:90% there'll be no moral hazard. 00:44:58.435 --> 00:44:59.560 align:middle line:90% So there's a simple answer. 00:44:59.560 --> 00:45:02.580 align:middle line:84% So for example, what if we said we're going to, for sure, 00:45:02.580 --> 00:45:05.280 align:middle line:84% give big benefits to people who are blind, who have lost an arm? 00:45:05.280 --> 00:45:07.100 align:middle line:84% Ignore my nub city, horrible example 00:45:07.100 --> 00:45:09.500 align:middle line:90% from the previous lecture. 00:45:09.500 --> 00:45:11.340 align:middle line:84% In that case, you might think, look, 00:45:11.340 --> 00:45:12.720 align:middle line:90% that is beyond people's control. 00:45:12.720 --> 00:45:15.095 align:middle line:84% They're not blinding themselves or cutting off their arms 00:45:15.095 --> 00:45:15.960 align:middle line:90% to get benefits. 00:45:15.960 --> 00:45:18.080 align:middle line:84% So that's an OK way to distribute. 00:45:18.080 --> 00:45:20.900 align:middle line:84% I'm not worried about moral hazard. 00:45:20.900 --> 00:45:23.100 align:middle line:84% So you might think that basically, 00:45:23.100 --> 00:45:26.780 align:middle line:84% an advantage, a third tool, besides g and tau, 00:45:26.780 --> 00:45:29.660 align:middle line:84% is using categorical restrictions 00:45:29.660 --> 00:45:34.350 align:middle line:90% as a way of redistributing. 00:45:34.350 --> 00:45:39.030 align:middle line:84% So basically, this comes the idea of targeting. 00:45:39.030 --> 00:45:41.550 align:middle line:84% The idea is, by targeting our dollars, 00:45:41.550 --> 00:45:47.750 align:middle line:84% we can spend them more effectively to meet these goals. 00:45:47.750 --> 00:45:50.190 align:middle line:84% So what would make a good targeting device? 00:45:50.190 --> 00:45:52.070 align:middle line:90% Well, it's two things. 00:45:52.070 --> 00:45:55.643 align:middle line:84% First of all, it's something you can't change. 00:45:55.643 --> 00:45:57.310 align:middle line:84% And second of all, it's something that's 00:45:57.310 --> 00:45:59.470 align:middle line:90% correlated with being poor. 00:45:59.470 --> 00:46:00.850 align:middle line:90% Let's not forget the second one. 00:46:00.850 --> 00:46:03.670 align:middle line:84% For example, if I said I'm going to set up a new program that's 00:46:03.670 --> 00:46:06.150 align:middle line:84% going to be $1 million to everyone who is a Fortune 500 00:46:06.150 --> 00:46:09.050 align:middle line:84% CEO last year, that would meet the first condition. 00:46:09.050 --> 00:46:10.430 align:middle line:90% They can't change that. 00:46:10.430 --> 00:46:12.350 align:middle line:84% It wouldn't be the second condition. 00:46:12.350 --> 00:46:14.495 align:middle line:84% That wouldn't be a very good transfer program. 00:46:14.495 --> 00:46:16.370 align:middle line:84% So you need both, something you can't change, 00:46:16.370 --> 00:46:20.950 align:middle line:84% but also, it's correlated with earnings ability. 00:46:20.950 --> 00:46:21.930 align:middle line:90% That's your ultimate. 00:46:21.930 --> 00:46:23.430 align:middle line:84% That's when something like blindness 00:46:23.430 --> 00:46:27.930 align:middle line:84% might work, or disability, or other things. 00:46:27.930 --> 00:46:30.750 align:middle line:84% Now the major targeting mechanism 00:46:30.750 --> 00:46:33.050 align:middle line:84% used by the US government for cash transfers, 00:46:33.050 --> 00:46:36.610 align:middle line:84% for many decades, was single motherhood. 00:46:36.610 --> 00:46:39.130 align:middle line:84% The idea was, single motherhood, for sure, 00:46:39.130 --> 00:46:42.390 align:middle line:84% met the second condition, which is that single mothers are much, 00:46:42.390 --> 00:46:44.487 align:middle line:90% much poorer. 00:46:44.487 --> 00:46:46.570 align:middle line:84% The poverty rate for single female-headed families 00:46:46.570 --> 00:46:48.690 align:middle line:84% with children is almost 30%, which 00:46:48.690 --> 00:46:53.570 align:middle line:84% is about triple the poverty rate for two-headed families 00:46:53.570 --> 00:46:55.390 align:middle line:90% with children. 00:46:55.390 --> 00:46:56.810 align:middle line:90% So they're certainly poor. 00:46:56.810 --> 00:46:59.030 align:middle line:84% The question is, is it something you can change? 00:46:59.030 --> 00:47:02.530 align:middle line:90% Well, clearly, you can. 00:47:02.530 --> 00:47:05.150 align:middle line:84% You're not innately a single mom. 00:47:05.150 --> 00:47:08.290 align:middle line:84% That's the result of some choices and some other things. 00:47:08.290 --> 00:47:14.650 align:middle line:84% So the question is, how big is the moral hazard from targeting 00:47:14.650 --> 00:47:15.677 align:middle line:90% single motherhood? 00:47:15.677 --> 00:47:17.510 align:middle line:84% Here's what's really fascinating about this. 00:47:17.510 --> 00:47:18.490 align:middle line:90% Think about it. 00:47:18.490 --> 00:47:22.090 align:middle line:84% On the one hand, you gain efficiency 00:47:22.090 --> 00:47:24.530 align:middle line:84% by giving the money to people are poor, regardless 00:47:24.530 --> 00:47:25.970 align:middle line:90% of how hard they work or not. 00:47:25.970 --> 00:47:28.350 align:middle line:84% On the other hand, you create a new problem, 00:47:28.350 --> 00:47:31.190 align:middle line:84% which is creating potentially single mothers, 00:47:31.190 --> 00:47:34.110 align:middle line:90% and it's a cost of targeting. 00:47:34.110 --> 00:47:36.050 align:middle line:84% You change behavior to meet the target. 00:47:36.050 --> 00:47:38.277 align:middle line:84% There's a new distortion in our economy, 00:47:38.277 --> 00:47:40.110 align:middle line:84% and there's lots of evidence that growing up 00:47:40.110 --> 00:47:41.987 align:middle line:84% in a single-parent household is bad for you, 00:47:41.987 --> 00:47:43.570 align:middle line:84% not that there aren't success stories. 00:47:43.570 --> 00:47:46.195 align:middle line:84% But on average, people who grow up in a single-parent household 00:47:46.195 --> 00:47:48.495 align:middle line:84% do less well than people who grow up with two parents. 00:47:48.495 --> 00:47:49.870 align:middle line:84% So if we're suddenly creating all 00:47:49.870 --> 00:47:53.790 align:middle line:84% these single-parent households, that would be bad. 00:47:53.790 --> 00:47:57.070 align:middle line:84% And this is what led to one of the most fundamental debates 00:47:57.070 --> 00:48:00.030 align:middle line:84% in welfare reform policy-- it dominated the period when I was 00:48:00.030 --> 00:48:01.070 align:middle line:90% a student-- 00:48:01.070 --> 00:48:04.070 align:middle line:84% which was a famous conservative sociologist 00:48:04.070 --> 00:48:06.150 align:middle line:90% called Charles Murray. 00:48:06.150 --> 00:48:09.350 align:middle line:84% And Charles Murray went to the data 00:48:09.350 --> 00:48:11.790 align:middle line:90% and noticed the following trend. 00:48:11.790 --> 00:48:15.930 align:middle line:84% He noticed that until-- well, actually, you can do this. 00:48:15.930 --> 00:48:16.910 align:middle line:90% You got to be honest. 00:48:16.910 --> 00:48:22.350 align:middle line:84% Take Figure 17.6 and hold your hand. 00:48:22.350 --> 00:48:28.650 align:middle line:84% Cut it off at 1979, or 1982, or something like that. 00:48:28.650 --> 00:48:34.850 align:middle line:84% The red line is the size of government transfers. 00:48:34.850 --> 00:48:37.570 align:middle line:84% The green line is single motherhood rate. 00:48:37.570 --> 00:48:40.910 align:middle line:84% What you see is, as government transfers fell-- 00:48:40.910 --> 00:48:47.770 align:middle line:90% 00:48:47.770 --> 00:48:49.590 align:middle line:84% actually, no, it doesn't even work there. 00:48:49.590 --> 00:48:52.615 align:middle line:90% It's not even very good. 00:48:52.615 --> 00:48:53.990 align:middle line:84% You can't even see in this graph. 00:48:53.990 --> 00:48:56.770 align:middle line:84% The bottom line is, for some period of time, 00:48:56.770 --> 00:49:00.633 align:middle line:84% he argued that as government transfers rose, 00:49:00.633 --> 00:49:02.050 align:middle line:84% single motherhood rose, and it was 00:49:02.050 --> 00:49:04.290 align:middle line:84% true over some small period of time. 00:49:04.290 --> 00:49:09.610 align:middle line:84% So basically, he was arguing that more transfers were 00:49:09.610 --> 00:49:11.110 align:middle line:90% causing more single motherhood. 00:49:11.110 --> 00:49:13.735 align:middle line:84% Of course, you can see the time series looks like, if anything, 00:49:13.735 --> 00:49:14.610 align:middle line:90% the opposite. 00:49:14.610 --> 00:49:16.570 align:middle line:84% And in fact, there's now been hundreds 00:49:16.570 --> 00:49:19.890 align:middle line:84% of studies which have shown that single motherhood does not 00:49:19.890 --> 00:49:23.610 align:middle line:84% respond to the generosity of cash welfare benefits. 00:49:23.610 --> 00:49:25.810 align:middle line:84% Now if you ask many politicians, they'll 00:49:25.810 --> 00:49:28.750 align:middle line:84% claim it's like a law of nature, that if you give cash 00:49:28.750 --> 00:49:30.790 align:middle line:84% to single moms, you'll create more single moms. 00:49:30.790 --> 00:49:32.570 align:middle line:84% But in fact, that's an empirical question. 00:49:32.570 --> 00:49:34.030 align:middle line:84% The answer is, we've answered it, 00:49:34.030 --> 00:49:37.650 align:middle line:84% which is tying cash benefits to single motherhood, 00:49:37.650 --> 00:49:41.030 align:middle line:84% does not induce more single motherhood. 00:49:41.030 --> 00:49:43.630 align:middle line:84% So it seems like single motherhood is actually 00:49:43.630 --> 00:49:47.070 align:middle line:90% a pretty good targeting device. 00:49:47.070 --> 00:49:50.650 align:middle line:84% It doesn't really respond to being targeted, 00:49:50.650 --> 00:49:53.670 align:middle line:84% and it identifies those who are low income. 00:49:53.670 --> 00:49:56.190 align:middle line:84% So that's an example of how we think about using 00:49:56.190 --> 00:49:58.450 align:middle line:90% categorical benefits. 00:49:58.450 --> 00:50:03.870 align:middle line:84% So basically, a very popular decision was, at one point-- 00:50:03.870 --> 00:50:06.070 align:middle line:84% as I, said people can actually get TANF, 00:50:06.070 --> 00:50:07.430 align:middle line:90% even if they're married. 00:50:07.430 --> 00:50:08.930 align:middle line:90% Very few people do, but you can. 00:50:08.930 --> 00:50:09.850 align:middle line:90% That was very popular. 00:50:09.850 --> 00:50:12.428 align:middle line:84% Ooh, we'll stop creating all these single mothers 00:50:12.428 --> 00:50:12.970 align:middle line:90% by giving it. 00:50:12.970 --> 00:50:14.390 align:middle line:84% What they didn't realize is, they didn't 00:50:14.390 --> 00:50:15.598 align:middle line:90% stop creating single mothers. 00:50:15.598 --> 00:50:17.990 align:middle line:84% They just gave out a lot more cash. 00:50:17.990 --> 00:50:21.610 align:middle line:84% And so it was like moving to lowering that tax rate. 00:50:21.610 --> 00:50:24.800 align:middle line:84% You didn't achieve all your goals. 00:50:24.800 --> 00:50:28.180 align:middle line:90% Questions about that? 00:50:28.180 --> 00:50:32.800 align:middle line:84% So that is one way you can lower the moral hazard, one way 00:50:32.800 --> 00:50:35.280 align:middle line:84% to solve this iron triangle problem. 00:50:35.280 --> 00:50:42.520 align:middle line:84% The second way is moving from cash to in-kind transfers. 00:50:42.520 --> 00:50:44.320 align:middle line:84% Moving from cash in-kind transfers 00:50:44.320 --> 00:50:49.060 align:middle line:84% also helps you solve this problem in a more indirect way. 00:50:49.060 --> 00:50:51.360 align:middle line:84% With categorical welfare, we do targeting. 00:50:51.360 --> 00:50:54.520 align:middle line:84% We find those who really need help and give money to them. 00:50:54.520 --> 00:50:56.520 align:middle line:84% With in-kind benefits, we do what 00:50:56.520 --> 00:51:03.280 align:middle line:84% we call self-selection, which is that we 00:51:03.280 --> 00:51:07.240 align:middle line:84% get those who need the benefits to identify themselves. 00:51:07.240 --> 00:51:08.760 align:middle line:90% And why is that true? 00:51:08.760 --> 00:51:10.442 align:middle line:84% That's because everyone wants cash, 00:51:10.442 --> 00:51:12.400 align:middle line:84% but not everyone wants to go to a soup kitchen. 00:51:12.400 --> 00:51:17.040 align:middle line:84% But if I give away cash, everyone's going to claim it. 00:51:17.040 --> 00:51:19.360 align:middle line:84% They're going to pretend they're poor and get it. 00:51:19.360 --> 00:51:21.360 align:middle line:84% But if I'm giving away a shitty public apartment 00:51:21.360 --> 00:51:23.720 align:middle line:84% or waiting outside in the rain for a soup kitchen, 00:51:23.720 --> 00:51:26.480 align:middle line:84% only people are really poor are going to actually do that, 00:51:26.480 --> 00:51:29.260 align:middle line:84% and that's a way to get them to reveal their true underlying 00:51:29.260 --> 00:51:33.300 align:middle line:84% ability, even if it's not stamped on their forehead. 00:51:33.300 --> 00:51:36.125 align:middle line:84% So this means, basically, the idea 00:51:36.125 --> 00:51:37.500 align:middle line:84% is, once again, we have two types 00:51:37.500 --> 00:51:44.900 align:middle line:84% of people, those that are innately skilled and those-- 00:51:44.900 --> 00:51:46.845 align:middle line:84% if you look at poor people, we have two types 00:51:46.845 --> 00:51:48.720 align:middle line:84% of poor people, innately skilled, lazy people 00:51:48.720 --> 00:51:51.820 align:middle line:84% and innately unskilled, hardworking people. 00:51:51.820 --> 00:51:53.907 align:middle line:84% The government would like to differentiate them. 00:51:53.907 --> 00:51:55.740 align:middle line:84% By giving cash, you can't differentiate them 00:51:55.740 --> 00:51:57.180 align:middle line:90% because everyone wants cash. 00:51:57.180 --> 00:52:00.060 align:middle line:84% But if you give them something which people really would only 00:52:00.060 --> 00:52:02.340 align:middle line:84% value if they had to have it, only the people 00:52:02.340 --> 00:52:06.680 align:middle line:84% who are really unskilled don't have a choice will take that. 00:52:06.680 --> 00:52:08.460 align:middle line:84% It's called an ordeal mechanism, so it's 00:52:08.460 --> 00:52:12.240 align:middle line:84% self-selection through an ordeal mechanism. 00:52:12.240 --> 00:52:15.792 align:middle line:90% 00:52:15.792 --> 00:52:17.000 align:middle line:90% I'm going to illustrate this. 00:52:17.000 --> 00:52:20.660 align:middle line:84% This is a hard concept because what it basically says, 00:52:20.660 --> 00:52:25.400 align:middle line:84% is you can make people better off by making them worse off. 00:52:25.400 --> 00:52:27.840 align:middle line:84% You make people better off by making them worse off. 00:52:27.840 --> 00:52:29.140 align:middle line:90% Let me explain what I mean. 00:52:29.140 --> 00:52:30.920 align:middle line:90% We'll do a simple example. 00:52:30.920 --> 00:52:36.560 align:middle line:84% Imagine that the government wants to set up a soup kitchen, 00:52:36.560 --> 00:52:41.120 align:middle line:84% and it has to decide how much to spend on hiring workers 00:52:41.120 --> 00:52:42.570 align:middle line:90% to staff the soup kitchen. 00:52:42.570 --> 00:52:44.320 align:middle line:84% And the more workers it hires, the shorter 00:52:44.320 --> 00:52:47.040 align:middle line:90% the wait to get in to get soup. 00:52:47.040 --> 00:52:49.560 align:middle line:84% So right now, the government's got this big line 00:52:49.560 --> 00:52:51.920 align:middle line:84% outside the soup kitchen, and it's got to decide, 00:52:51.920 --> 00:52:53.420 align:middle line:84% is that a good thing or a bad thing? 00:52:53.420 --> 00:52:54.837 align:middle line:84% Well, it's a bad thing that people 00:52:54.837 --> 00:52:57.840 align:middle line:84% will wait in rain for their soup, but it has an advantage. 00:52:57.840 --> 00:52:59.840 align:middle line:90% It's important to recognize. 00:52:59.840 --> 00:53:01.360 align:middle line:84% So imagine that there's two types 00:53:01.360 --> 00:53:04.280 align:middle line:84% of people, low potential earnings 00:53:04.280 --> 00:53:06.440 align:middle line:90% and high potential earnings. 00:53:06.440 --> 00:53:10.000 align:middle line:84% The utility function for low potential earnings person, u sub 00:53:10.000 --> 00:53:14.800 align:middle line:90% l is 240. 00:53:14.800 --> 00:53:22.620 align:middle line:84% U sub l is 240s minus w, where's s is the number of bowls of soup 00:53:22.620 --> 00:53:27.540 align:middle line:84% and w is the amount of minutes they wait for soup. 00:53:27.540 --> 00:53:33.860 align:middle line:84% The utility of a high-potential earnings person is 120s minus 00:53:33.860 --> 00:53:34.800 align:middle line:90% 2w. 00:53:34.800 --> 00:53:37.340 align:middle line:90% 00:53:37.340 --> 00:53:39.545 align:middle line:90% Why does it make sense? 00:53:39.545 --> 00:53:41.920 align:middle line:84% Why do these two differences make sense in this equation? 00:53:41.920 --> 00:53:43.878 align:middle line:84% First of all, why would a high-potential person 00:53:43.878 --> 00:53:47.100 align:middle line:84% have a lower wait than the amount of soup they get? 00:53:47.100 --> 00:53:48.500 align:middle line:90% What is this showing? 00:53:48.500 --> 00:53:49.440 align:middle line:90% What is illustrating? 00:53:49.440 --> 00:53:51.540 align:middle line:90% Why is that? 00:53:51.540 --> 00:53:53.100 align:middle line:90% Yeah? 00:53:53.100 --> 00:53:55.575 align:middle line:84% Well, they have a lower marginal utility of consumption, 00:53:55.575 --> 00:53:57.700 align:middle line:84% because they can always get more food if they want. 00:53:57.700 --> 00:53:58.860 align:middle line:90% They're high ability. 00:53:58.860 --> 00:54:02.560 align:middle line:84% So they don't need the soup as much. 00:54:02.560 --> 00:54:04.605 align:middle line:84% Why do we subtract 2 times the number of minutes 00:54:04.605 --> 00:54:06.480 align:middle line:84% and only 1 times 100 minutes for that person? 00:54:06.480 --> 00:54:06.940 align:middle line:90% Enoch? 00:54:06.940 --> 00:54:07.620 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:54:07.620 --> 00:54:08.640 align:middle line:90% PROFESSOR: Opportunity cost. 00:54:08.640 --> 00:54:10.340 align:middle line:84% If you're high skilled, you have a higher opportunity cost 00:54:10.340 --> 00:54:12.340 align:middle line:84% waiting around outside the soup kitchen. 00:54:12.340 --> 00:54:14.940 align:middle line:84% So this is a totally plausible comparison 00:54:14.940 --> 00:54:19.092 align:middle line:84% of two types, the type that really needs the Soup 00:54:19.092 --> 00:54:21.550 align:middle line:84% but doesn't mind waiting and the type doesn't need the soup 00:54:21.550 --> 00:54:24.550 align:middle line:90% as much and minds waiting more. 00:54:24.550 --> 00:54:27.350 align:middle line:84% And imagine that we're going to find social welfare as simply 00:54:27.350 --> 00:54:28.510 align:middle line:90% the sum of these. 00:54:28.510 --> 00:54:30.470 align:middle line:84% Adding those up as social welfare. 00:54:30.470 --> 00:54:32.970 align:middle line:84% We're not going to put more weight on the poor and the rich. 00:54:32.970 --> 00:54:37.150 align:middle line:84% We're just going to say we're going to add them together. 00:54:37.150 --> 00:54:39.990 align:middle line:84% Now imagine that the government implements 00:54:39.990 --> 00:54:42.070 align:middle line:90% a system with no waiting. 00:54:42.070 --> 00:54:43.838 align:middle line:90% W equals 0. 00:54:43.838 --> 00:54:46.130 align:middle line:84% And imagine a system with no waiting to make life easy. 00:54:46.130 --> 00:54:49.070 align:middle line:84% Imagine that each group gets the same amount 00:54:49.070 --> 00:54:50.430 align:middle line:90% of soup but no waiting. 00:54:50.430 --> 00:54:52.190 align:middle line:90% Just imagine it's split. 00:54:52.190 --> 00:54:54.950 align:middle line:90% It's just allocated randomly. 00:54:54.950 --> 00:55:03.710 align:middle line:84% Well, if w equals 0, u sub l is 240 because they get one soup. 00:55:03.710 --> 00:55:06.310 align:middle line:84% Let's say there's two bowls of soup. 00:55:06.310 --> 00:55:07.630 align:middle line:90% They get one soup. 00:55:07.630 --> 00:55:13.630 align:middle line:84% And u sub h is 120 for a total social welfare 360. 00:55:13.630 --> 00:55:17.250 align:middle line:84% So if there's no waiting time, they each get a bowl of soup 00:55:17.250 --> 00:55:20.990 align:middle line:84% and social welfare is 360, because this term drops out 00:55:20.990 --> 00:55:23.120 align:middle line:90% because there's no waiting. 00:55:23.120 --> 00:55:24.870 align:middle line:84% Now imagine the government, instead, said, 00:55:24.870 --> 00:55:27.050 align:middle line:84% I'm not going to hire so many workers, 00:55:27.050 --> 00:55:29.450 align:middle line:84% and I'm going to make people wait outside for 61 00:55:29.450 --> 00:55:31.170 align:middle line:90% minutes before they get soup. 00:55:31.170 --> 00:55:32.510 align:middle line:90% I'm going to literally mandate-- 00:55:32.510 --> 00:55:33.510 align:middle line:90% I'm going to be so mean. 00:55:33.510 --> 00:55:35.490 align:middle line:84% I'm going to mandate you have to stand out-- even if there's 00:55:35.490 --> 00:55:37.490 align:middle line:84% no one inside, just stand outside in the fucking 00:55:37.490 --> 00:55:41.090 align:middle line:84% rain for 61 minutes and wait to get your soup. 00:55:41.090 --> 00:55:42.130 align:middle line:90% Awful, right? 00:55:42.130 --> 00:55:46.080 align:middle line:90% Except think about what it does. 00:55:46.080 --> 00:55:49.850 align:middle line:84% At a 61-minute wait, the high-ability person no longer 00:55:49.850 --> 00:55:51.490 align:middle line:90% wants soup. 00:55:51.490 --> 00:55:52.710 align:middle line:90% The utility goes negative. 00:55:52.710 --> 00:55:54.650 align:middle line:90% They're not going to wait. 00:55:54.650 --> 00:56:02.290 align:middle line:84% So with a 61-minute wait, u sub h equals 0. 00:56:02.290 --> 00:56:03.110 align:middle line:90% What's u sub l? 00:56:03.110 --> 00:56:10.370 align:middle line:84% Well, now they get two bowls of soup, so they get 480 minus 61, 00:56:10.370 --> 00:56:13.550 align:middle line:90% which equals 419. 00:56:13.550 --> 00:56:19.390 align:middle line:84% So total social welfare has risen from 360 to 419. 00:56:19.390 --> 00:56:23.590 align:middle line:84% We've made poor people better off by making them worse off. 00:56:23.590 --> 00:56:28.390 align:middle line:84% It's a bizarre concept but really, really interesting. 00:56:28.390 --> 00:56:29.530 align:middle line:90% Why is that? 00:56:29.530 --> 00:56:30.410 align:middle line:90% What's the key here? 00:56:30.410 --> 00:56:31.750 align:middle line:84% It's because we have a fixed government budget. 00:56:31.750 --> 00:56:33.750 align:middle line:84% With an infinite budget, it wouldn't matter. 00:56:33.750 --> 00:56:35.470 align:middle line:84% In the fixed government budget, the more 00:56:35.470 --> 00:56:38.270 align:middle line:84% you can target to those who really need it, 00:56:38.270 --> 00:56:39.767 align:middle line:90% the more efficient. 00:56:39.767 --> 00:56:41.350 align:middle line:84% We could target it by having something 00:56:41.350 --> 00:56:43.650 align:middle line:84% stamped on people's heads or single motherhood, 00:56:43.650 --> 00:56:45.590 align:middle line:84% or we could target it by letting people reveal 00:56:45.590 --> 00:56:47.830 align:middle line:90% whether they need it or not. 00:56:47.830 --> 00:56:50.870 align:middle line:84% And one of the reasons we have in-kind benefits 00:56:50.870 --> 00:56:56.970 align:middle line:84% is to let people reveal how much they value the good. 00:56:56.970 --> 00:57:00.830 align:middle line:90% 00:57:00.830 --> 00:57:04.150 align:middle line:84% So for example, there's a really cool study 00:57:04.150 --> 00:57:10.070 align:middle line:84% I talk about in the book, which is what happened when a state-- 00:57:10.070 --> 00:57:11.820 align:middle line:84% the government ran an experiment. 00:57:11.820 --> 00:57:14.070 align:middle line:84% It was about taking care of old people in their homes. 00:57:14.070 --> 00:57:15.570 align:middle line:84% The government ran an experiment. 00:57:15.570 --> 00:57:18.810 align:middle line:84% They either gave people cash or they gave them 00:57:18.810 --> 00:57:21.210 align:middle line:90% a worker to work in their home. 00:57:21.210 --> 00:57:26.530 align:middle line:84% And what they found was, when they gave people the worker 00:57:26.530 --> 00:57:28.943 align:middle line:84% to work in their home, it cost a ton less, 00:57:28.943 --> 00:57:31.610 align:middle line:84% and the people who used it were the people who were really sick. 00:57:31.610 --> 00:57:32.270 align:middle line:90% [INAUDIBLE] cash. 00:57:32.270 --> 00:57:32.978 align:middle line:90% Everyone used it. 00:57:32.978 --> 00:57:34.890 align:middle line:90% It cost a ton more. 00:57:34.890 --> 00:57:37.665 align:middle line:84% It was much better targeted when they actually said, well, 00:57:37.665 --> 00:57:38.790 align:middle line:90% you got to have the worker. 00:57:38.790 --> 00:57:40.630 align:middle line:84% People were like, I'm not going to do that. 00:57:40.630 --> 00:57:41.750 align:middle line:84% I don't want someone in my house, 00:57:41.750 --> 00:57:43.708 align:middle line:84% whereas you gave them cash, they're like, sure, 00:57:43.708 --> 00:57:44.650 align:middle line:90% I'll take that cash. 00:57:44.650 --> 00:57:48.047 align:middle line:84% So basically, people were self-selecting in a way 00:57:48.047 --> 00:57:49.630 align:middle line:84% that allowed them to target the money. 00:57:49.630 --> 00:57:52.530 align:middle line:90% 00:57:52.530 --> 00:57:54.210 align:middle line:90% Questions about that? 00:57:54.210 --> 00:57:54.770 align:middle line:90% Yeah? 00:57:54.770 --> 00:58:00.170 align:middle line:84% AUDIENCE: [INAUDIBLE] I don't know if this is just a sample 00:58:00.170 --> 00:58:02.970 align:middle line:84% or whatever of people can't get to the soup kitchen. 00:58:02.970 --> 00:58:07.730 align:middle line:84% PROFESSOR: Well, once again, I'm not saying all soup kitchens 00:58:07.730 --> 00:58:09.430 align:middle line:90% in America should have a line. 00:58:09.430 --> 00:58:11.150 align:middle line:84% I'm saying this is the trade off, 00:58:11.150 --> 00:58:13.030 align:middle line:90% and you want to evaluate that. 00:58:13.030 --> 00:58:15.097 align:middle line:84% It may be that what you're saying 00:58:15.097 --> 00:58:17.430 align:middle line:84% is that maybe there's a big weight on w for these people 00:58:17.430 --> 00:58:18.690 align:middle line:90% because they can't get there. 00:58:18.690 --> 00:58:20.357 align:middle line:84% Well, then that would change the answer. 00:58:20.357 --> 00:58:23.830 align:middle line:84% If this was 2w, you wouldn't like this new policy. 00:58:23.830 --> 00:58:27.430 align:middle line:84% You'd be worse off under the new policy. 00:58:27.430 --> 00:58:29.210 align:middle line:84% The point is to illustrate the trade-off, 00:58:29.210 --> 00:58:31.990 align:middle line:84% And you're saying that's another issue of the trade-off. 00:58:31.990 --> 00:58:32.790 align:middle line:90% Yeah? 00:58:32.790 --> 00:58:35.630 align:middle line:84% AUDIENCE: How about increasing the cost 00:58:35.630 --> 00:58:39.350 align:middle line:84% of accessing a program with people 00:58:39.350 --> 00:58:43.730 align:middle line:84% to [INAUDIBLE] the population that you want [INAUDIBLE]. 00:58:43.730 --> 00:58:45.470 align:middle line:84% PROFESSOR: So basically, let me rephrase 00:58:45.470 --> 00:58:46.690 align:middle line:90% your question a little bit. 00:58:46.690 --> 00:58:50.708 align:middle line:84% The bottom line is that we talked about only partial 00:58:50.708 --> 00:58:52.750 align:middle line:84% take up, that some people don't take up benefits. 00:58:52.750 --> 00:58:54.390 align:middle line:90% That could be good. 00:58:54.390 --> 00:58:56.368 align:middle line:84% That could be illustrating who really needs it. 00:58:56.368 --> 00:58:57.910 align:middle line:84% And in fact, some grad student at MIT 00:58:57.910 --> 00:58:59.790 align:middle line:84% has a really cool new study where 00:58:59.790 --> 00:59:02.455 align:middle line:84% they look across seven different transfer programs. 00:59:02.455 --> 00:59:03.830 align:middle line:84% All the programs target you based 00:59:03.830 --> 00:59:05.350 align:middle line:90% on what your income is today. 00:59:05.350 --> 00:59:06.510 align:middle line:84% But of course, people's income today 00:59:06.510 --> 00:59:08.500 align:middle line:84% can be very different than income over their life. 00:59:08.500 --> 00:59:10.375 align:middle line:84% We really want to care about people who are-- 00:59:10.375 --> 00:59:11.555 align:middle line:90% so a student has low income. 00:59:11.555 --> 00:59:13.180 align:middle line:84% A student at MIT might have low income. 00:59:13.180 --> 00:59:14.580 align:middle line:84% We don't really want to distribute to them. 00:59:14.580 --> 00:59:15.260 align:middle line:90% Sorry, guys. 00:59:15.260 --> 00:59:16.360 align:middle line:90% You're going to be fine. 00:59:16.360 --> 00:59:20.620 align:middle line:84% What they find is that people who take up, 00:59:20.620 --> 00:59:23.760 align:middle line:84% at any given point in time are the same people who end up 00:59:23.760 --> 00:59:25.375 align:middle line:90% with lower lifetime income. 00:59:25.375 --> 00:59:27.000 align:middle line:84% The people you want to take the program 00:59:27.000 --> 00:59:28.380 align:middle line:84% are the ones who choose to take it. 00:59:28.380 --> 00:59:30.588 align:middle line:84% The ones who don't take it are the ones whose incomes 00:59:30.588 --> 00:59:31.480 align:middle line:90% are temporarily low. 00:59:31.480 --> 00:59:33.800 align:middle line:84% But actually, their income is going to be higher later. 00:59:33.800 --> 00:59:35.620 align:middle line:84% So that's a reason why, in some sense, 00:59:35.620 --> 00:59:38.080 align:middle line:84% incomplete take up might be good. 00:59:38.080 --> 00:59:41.960 align:middle line:84% That's another argument for in-kind transfers. 00:59:41.960 --> 00:59:43.140 align:middle line:90% Yeah, Enoch? 00:59:43.140 --> 00:59:44.640 align:middle line:84% AUDIENCE: Why doesn't the government 00:59:44.640 --> 00:59:47.760 align:middle line:84% not make it as hard as possible to get these programs to just 00:59:47.760 --> 00:59:50.120 align:middle line:84% make it increasingly more difficult? 00:59:50.120 --> 00:59:51.960 align:middle line:90% PROFESSOR: Because of what-- 00:59:51.960 --> 00:59:52.740 align:middle line:90% Sophia, right? 00:59:52.740 --> 00:59:54.720 align:middle line:84% Because of what Sophia pointed out, 00:59:54.720 --> 00:59:56.500 align:middle line:84% which is that basically, there's a cost. 00:59:56.500 --> 00:59:57.740 align:middle line:84% That ordeal mechanism has a cost. 00:59:57.740 --> 00:59:58.948 align:middle line:90% They got to wait in the rain. 00:59:58.948 --> 00:59:59.920 align:middle line:90% That sucks. 00:59:59.920 --> 01:00:03.240 align:middle line:84% So you've got to trade off how bad the cost is 01:00:03.240 --> 01:00:05.120 align:middle line:90% against the targeting benefit. 01:00:05.120 --> 01:00:10.180 align:middle line:84% And the sweet spot is one where it's enough of a pain 01:00:10.180 --> 01:00:12.660 align:middle line:84% to deter the high ability, but it's still very valuable 01:00:12.660 --> 01:00:14.540 align:middle line:84% for the low ability, as in this example. 01:00:14.540 --> 01:00:17.140 align:middle line:84% But like I said, if you change this to 2w or change this 01:00:17.140 --> 01:00:19.380 align:middle line:90% to 120, it doesn't work. 01:00:19.380 --> 01:00:21.900 align:middle line:84% So really, that's the trade-off, the cost 01:00:21.900 --> 01:00:25.900 align:middle line:84% of the ordeal versus the benefit of targeting to the group that 01:00:25.900 --> 01:00:27.180 align:middle line:90% really needs it. 01:00:27.180 --> 01:00:28.880 align:middle line:90% And that's the trade-off here. 01:00:28.880 --> 01:00:29.420 align:middle line:90% Yeah? 01:00:29.420 --> 01:00:31.420 align:middle line:84% AUDIENCE: If w is too high, then you 01:00:31.420 --> 01:00:33.900 align:middle line:84% get social welfare that's below the-- 01:00:33.900 --> 01:00:34.767 align:middle line:90% PROFESSOR: Yes. 01:00:34.767 --> 01:00:37.100 align:middle line:84% AUDIENCE: Is there evidence that the government's target 01:00:37.100 --> 01:00:40.457 align:middle line:84% [INAUDIBLE] the perfect amount of pain-- 01:00:40.457 --> 01:00:42.040 align:middle line:84% PROFESSOR: That's an awesome question. 01:00:42.040 --> 01:00:44.080 align:middle line:90% It's an awesome question. 01:00:44.080 --> 01:00:46.940 align:middle line:84% basically, there's the fundamental challenge. 01:00:46.940 --> 01:00:48.920 align:middle line:84% Here's the problem with that whole study. 01:00:48.920 --> 01:00:51.663 align:middle line:84% It comes back to poverty line measurement, which is-- 01:00:51.663 --> 01:00:53.580 align:middle line:84% let's go back to the poverty line measurement. 01:00:53.580 --> 01:00:55.140 align:middle line:84% Ideally, if we measure poverty, we'd 01:00:55.140 --> 01:00:56.807 align:middle line:84% include what you're getting in Medicaid. 01:00:56.807 --> 01:00:59.390 align:middle line:84% But then I have to ask what it's worth to you to get Medicaid. 01:00:59.390 --> 01:01:00.360 align:middle line:90% I want a dollar metric. 01:01:00.360 --> 01:01:02.180 align:middle line:84% Well, how the hell do I know what it's worth to you 01:01:02.180 --> 01:01:03.040 align:middle line:90% to get Medicaid? 01:01:03.040 --> 01:01:04.890 align:middle line:84% You could use the average cost of Medicaid, 01:01:04.890 --> 01:01:06.640 align:middle line:84% but it's worth a lot more to a sick person 01:01:06.640 --> 01:01:08.040 align:middle line:90% than a healthy person. 01:01:08.040 --> 01:01:10.400 align:middle line:84% It's the same thing here, which is, in some sense, 01:01:10.400 --> 01:01:12.480 align:middle line:84% to really do this right. we need to not only know 01:01:12.480 --> 01:01:14.860 align:middle line:84% what the wait on w is for things like the wheelchair bound. 01:01:14.860 --> 01:01:16.443 align:middle line:84% We need to know how much they actually 01:01:16.443 --> 01:01:18.800 align:middle line:90% value this relative to cash. 01:01:18.800 --> 01:01:20.660 align:middle line:84% And this will come back to the UBI debate 01:01:20.660 --> 01:01:23.620 align:middle line:84% we'll talk about next time, which is, in some sense, 01:01:23.620 --> 01:01:25.880 align:middle line:84% we may be outsmarting ourselves, doing 01:01:25.880 --> 01:01:27.020 align:middle line:90% all these in-kind benefits. 01:01:27.020 --> 01:01:28.320 align:middle line:84% Because the truth is, it's just better 01:01:28.320 --> 01:01:30.112 align:middle line:84% to give people cash and let them figure out 01:01:30.112 --> 01:01:31.440 align:middle line:90% what they want to spend it on. 01:01:31.440 --> 01:01:33.240 align:middle line:84% Now that said, there are a number 01:01:33.240 --> 01:01:36.720 align:middle line:84% of other advantages of in-kind programs. 01:01:36.720 --> 01:01:40.160 align:middle line:84% Another advantage that's a really interesting one 01:01:40.160 --> 01:01:41.980 align:middle line:90% is that it affects the market. 01:01:41.980 --> 01:01:44.840 align:middle line:84% So in this cool project in Mexico, 01:01:44.840 --> 01:01:47.000 align:middle line:84% where they give some people cash, 01:01:47.000 --> 01:01:50.840 align:middle line:84% some communities got a cash transfer and some communities 01:01:50.840 --> 01:01:53.502 align:middle line:90% got an in-kind transfer of food. 01:01:53.502 --> 01:01:54.960 align:middle line:84% What they found was the communities 01:01:54.960 --> 01:01:59.580 align:middle line:84% that got the in-kind transfer, the same size, were better off. 01:01:59.580 --> 01:02:00.080 align:middle line:90% Why? 01:02:00.080 --> 01:02:01.980 align:middle line:90% Because of the market for food. 01:02:01.980 --> 01:02:03.040 align:middle line:90% When you gave people-- 01:02:03.040 --> 01:02:03.540 align:middle line:90% Here's. 01:02:03.540 --> 01:02:05.380 align:middle line:84% The market for food-- quantity and price, 01:02:05.380 --> 01:02:08.140 align:middle line:84% you've got some initial equilibrium. 01:02:08.140 --> 01:02:11.900 align:middle line:84% When you gave people cash, what did that do? 01:02:11.900 --> 01:02:13.740 align:middle line:90% It raised the demand for food. 01:02:13.740 --> 01:02:15.900 align:middle line:90% So demand shifted. 01:02:15.900 --> 01:02:16.620 align:middle line:90% It raised demand. 01:02:16.620 --> 01:02:18.142 align:middle line:90% Demand shifted out. 01:02:18.142 --> 01:02:20.600 align:middle line:84% That led to a higher price for food, more quantity of food, 01:02:20.600 --> 01:02:22.300 align:middle line:90% but a higher price. 01:02:22.300 --> 01:02:25.840 align:middle line:84% When you give them in-kind, what does it do? 01:02:25.840 --> 01:02:27.340 align:middle line:90% It raises the supply. 01:02:27.340 --> 01:02:29.300 align:middle line:90% When you raise the supply-- 01:02:29.300 --> 01:02:33.165 align:middle line:84% when we lower supply, you also get more, but at a lower price. 01:02:33.165 --> 01:02:34.040 align:middle line:90% Why does that matter? 01:02:34.040 --> 01:02:36.940 align:middle line:84% Because some people aren't getting cash transfers. 01:02:36.940 --> 01:02:39.720 align:middle line:84% So in-kind had the benefit that only redistributed it. 01:02:39.720 --> 01:02:42.300 align:middle line:84% It lowered the price for everyone else, 01:02:42.300 --> 01:02:44.040 align:middle line:90% which is a cool feature of it. 01:02:44.040 --> 01:02:46.237 align:middle line:84% So that's another interesting feature of in-kind, 01:02:46.237 --> 01:02:48.320 align:middle line:84% is that it can have this market adjustment effect. 01:02:48.320 --> 01:02:48.820 align:middle line:90% Yeah? 01:02:48.820 --> 01:02:53.260 align:middle line:84% AUDIENCE: [INAUDIBLE] they got food from food vendors 01:02:53.260 --> 01:02:55.000 align:middle line:84% within the community [INAUDIBLE]. 01:02:55.000 --> 01:02:57.307 align:middle line:90% 01:02:57.307 --> 01:02:58.140 align:middle line:90% PROFESSOR: Well, no. 01:02:58.140 --> 01:02:59.660 align:middle line:84% You're right, because there's also 01:02:59.660 --> 01:03:02.950 align:middle line:84% an effect on the producers themselves. 01:03:02.950 --> 01:03:04.790 align:middle line:84% So you'd have to consider there's 01:03:04.790 --> 01:03:05.950 align:middle line:84% a general equilibrium where you consider 01:03:05.950 --> 01:03:06.850 align:middle line:90% the producers themselves. 01:03:06.850 --> 01:03:08.740 align:middle line:84% This is a diagram that works with the food 01:03:08.740 --> 01:03:10.150 align:middle line:84% that all comes from other people. 01:03:10.150 --> 01:03:11.650 align:middle line:84% Once in the community, you then have 01:03:11.650 --> 01:03:15.030 align:middle line:84% to wait for effect of the producers in the community. 01:03:15.030 --> 01:03:17.030 align:middle line:84% And then basically, the other thing 01:03:17.030 --> 01:03:19.750 align:middle line:84% that might be good about in-kind transfers 01:03:19.750 --> 01:03:22.670 align:middle line:84% is that sometimes, cash can't buy things. 01:03:22.670 --> 01:03:25.870 align:middle line:84% So for example, let's think about giving you health 01:03:25.870 --> 01:03:29.190 align:middle line:84% insurance versus giving you, say, $10,000. 01:03:29.190 --> 01:03:31.830 align:middle line:84% Well, especially, let's think before the ACA. 01:03:31.830 --> 01:03:34.753 align:middle line:84% Before the ACA, if I gave a sick person $10,000, 01:03:34.753 --> 01:03:36.170 align:middle line:84% they couldn't do anything with it. 01:03:36.170 --> 01:03:37.630 align:middle line:84% They couldn't get the health insurance 01:03:37.630 --> 01:03:39.463 align:middle line:84% because they could be discriminated against. 01:03:39.463 --> 01:03:41.673 align:middle line:84% If I gave them health insurance, they got it. 01:03:41.673 --> 01:03:43.590 align:middle line:84% So the other thing is, there may be situations 01:03:43.590 --> 01:03:45.715 align:middle line:84% where the government is essentially fixing a market 01:03:45.715 --> 01:03:49.870 align:middle line:84% failure by redistributing in kind rather than with cash. 01:03:49.870 --> 01:03:52.910 align:middle line:84% So that's another argument for in-kind. 01:03:52.910 --> 01:03:55.590 align:middle line:84% And then the final argument for in-kind 01:03:55.590 --> 01:03:57.290 align:middle line:84% is the one that dominates all of them. 01:03:57.290 --> 01:03:59.770 align:middle line:84% I hate it, but it's just politics. 01:03:59.770 --> 01:04:00.570 align:middle line:90% It's just that. 01:04:00.570 --> 01:04:02.737 align:middle line:84% Politicians are more sympathetic about giving people 01:04:02.737 --> 01:04:04.250 align:middle line:90% food and medical care than cash. 01:04:04.250 --> 01:04:06.792 align:middle line:84% That's the main reason why we give most of the money in kind, 01:04:06.792 --> 01:04:07.530 align:middle line:90% I'm sorry to say. 01:04:07.530 --> 01:04:10.090 align:middle line:84% All the fun economics is just fun economics. 01:04:10.090 --> 01:04:12.690 align:middle line:84% But the truth is, from a chapter 9 perspective, 01:04:12.690 --> 01:04:14.690 align:middle line:84% the main reason politicians get benefits in kind 01:04:14.690 --> 01:04:18.910 align:middle line:84% is because they don't trust people with the cash. 01:04:18.910 --> 01:04:19.410 align:middle line:90% Yeah? 01:04:19.410 --> 01:04:22.030 align:middle line:84% AUDIENCE: Would a reimbursement be the same thing as cash? 01:04:22.030 --> 01:04:23.355 align:middle line:90% PROFESSOR: A reimbursement? 01:04:23.355 --> 01:04:24.730 align:middle line:84% AUDIENCE: Like, a rebate program. 01:04:24.730 --> 01:04:25.490 align:middle line:90% PROFESSOR: Yeah. 01:04:25.490 --> 01:04:27.210 align:middle line:90% Yeah. 01:04:27.210 --> 01:04:29.470 align:middle line:84% Actually, let's go back to chapter 10. 01:04:29.470 --> 01:04:31.133 align:middle line:90% That'd be like a matching grant. 01:04:31.133 --> 01:04:32.550 align:middle line:84% You can either give a block grant, 01:04:32.550 --> 01:04:33.930 align:middle line:90% which is just giving them cash. 01:04:33.930 --> 01:04:35.730 align:middle line:84% If you match the grant, you say, we're 01:04:35.730 --> 01:04:37.850 align:middle line:84% going to rebate you as you buy it. 01:04:37.850 --> 01:04:41.090 align:middle line:84% That would be a middle ground, but that 01:04:41.090 --> 01:04:46.290 align:middle line:84% wouldn't have the a deal benefit of actually tying it to it. 01:04:46.290 --> 01:04:48.870 align:middle line:90% So fundamentally, we have this-- 01:04:48.870 --> 01:04:50.106 align:middle line:90% Yeah? 01:04:50.106 --> 01:04:52.250 align:middle line:84% AUDIENCE: [INAUDIBLE] because you 01:04:52.250 --> 01:04:55.330 align:middle line:84% argued that both the time and cash benefit [INAUDIBLE]. 01:04:55.330 --> 01:04:56.450 align:middle line:90% How do you decide? 01:04:56.450 --> 01:04:59.710 align:middle line:90% PROFESSOR: Great question. 01:04:59.710 --> 01:05:01.335 align:middle line:84% Actually, there's enormous disagreement 01:05:01.335 --> 01:05:03.502 align:middle line:84% in the economic profession, including between myself 01:05:03.502 --> 01:05:04.470 align:middle line:90% and my colleagues. 01:05:04.470 --> 01:05:05.860 align:middle line:84% So my colleague, Amy Finkelstein, 01:05:05.860 --> 01:05:08.110 align:middle line:84% believes we should stop giving people health insurance 01:05:08.110 --> 01:05:09.230 align:middle line:90% to give them cash. 01:05:09.230 --> 01:05:10.550 align:middle line:90% I disagree. 01:05:10.550 --> 01:05:12.910 align:middle line:84% I think that basically, there's enough imperfections 01:05:12.910 --> 01:05:14.410 align:middle line:84% in the health insurance decision. 01:05:14.410 --> 01:05:16.410 align:middle line:84% Her view is in a world where fully informed individuals, 01:05:16.410 --> 01:05:17.570 align:middle line:90% they'd rather have the cash. 01:05:17.570 --> 01:05:18.970 align:middle line:84% I think there's enough behavioral problems. 01:05:18.970 --> 01:05:20.845 align:middle line:84% We should give them health insurance instead. 01:05:20.845 --> 01:05:22.550 align:middle line:84% But she's smarter than I am, and that's 01:05:22.550 --> 01:05:23.970 align:middle line:90% an honest debate we're having. 01:05:23.970 --> 01:05:26.070 align:middle line:84% And so basically, I think there's a real debate. 01:05:26.070 --> 01:05:29.750 align:middle line:90% 01:05:29.750 --> 01:05:31.750 align:middle line:84% If I just summarized the consensus-- 01:05:31.750 --> 01:05:33.630 align:middle line:84% and I wouldn't want to be quizzed by my Nobel 01:05:33.630 --> 01:05:35.510 align:middle line:84% Prize-winning colleagues, Esther Duflo and Abhijit 01:05:35.510 --> 01:05:36.218 align:middle line:90% Banerjee on this. 01:05:36.218 --> 01:05:37.390 align:middle line:90% They're the experts. 01:05:37.390 --> 01:05:39.510 align:middle line:84% But my guess is, where the consensus today 01:05:39.510 --> 01:05:43.190 align:middle line:84% is that in developing countries, it's better to just give cash. 01:05:43.190 --> 01:05:47.330 align:middle line:90% 01:05:47.330 --> 01:05:49.550 align:middle line:84% There are different market failures and things 01:05:49.550 --> 01:05:52.350 align:middle line:84% that make it hard for the government 01:05:52.350 --> 01:05:54.230 align:middle line:84% to figure exactly how to give them the goods. 01:05:54.230 --> 01:05:57.690 align:middle line:84% In developed countries, I think there's more of an open debate. 01:05:57.690 --> 01:06:00.610 align:middle line:90% AUDIENCE: [INAUDIBLE] 01:06:00.610 --> 01:06:04.800 align:middle line:90% 01:06:04.800 --> 01:06:06.550 align:middle line:84% PROFESSOR: So that's the second mechanism. 01:06:06.550 --> 01:06:07.670 align:middle line:90% So once again, let's go back. 01:06:07.670 --> 01:06:09.470 align:middle line:84% We're trying to go around this iron triangle. 01:06:09.470 --> 01:06:10.770 align:middle line:84% One mechanism is trying to tie it 01:06:10.770 --> 01:06:12.145 align:middle line:84% to the people who really need, it 01:06:12.145 --> 01:06:14.268 align:middle line:84% like blind people and single mothers. 01:06:14.268 --> 01:06:16.810 align:middle line:84% The other is to give it in kind, so people will really reveal 01:06:16.810 --> 01:06:18.570 align:middle line:90% themselves when they need it. 01:06:18.570 --> 01:06:22.090 align:middle line:84% And then there's, finally, the easiest answer of all, which 01:06:22.090 --> 01:06:25.890 align:middle line:84% is, let's just increase people's options so they don't need it. 01:06:25.890 --> 01:06:28.090 align:middle line:84% Let's increase people's outside options. 01:06:28.090 --> 01:06:37.810 align:middle line:84% So for example, go to Figure 17.8, 01:06:37.810 --> 01:06:42.330 align:middle line:84% We have our old-style welfare program where people person 01:06:42.330 --> 01:06:44.170 align:middle line:90% x jumped on and person z does. 01:06:44.170 --> 01:06:46.050 align:middle line:90% I'm showing you person y here. 01:06:46.050 --> 01:06:49.290 align:middle line:90% And person y went to point d. 01:06:49.290 --> 01:06:51.170 align:middle line:90% That's the blue line. 01:06:51.170 --> 01:06:53.860 align:middle line:90% What if I raised your wage? 01:06:53.860 --> 01:06:59.128 align:middle line:84% What if people's wages increased from $15 an hour to $17.50? 01:06:59.128 --> 01:07:00.920 align:middle line:84% What does that do to the budget constraint? 01:07:00.920 --> 01:07:02.020 align:middle line:90% Well, it pivots it up. 01:07:02.020 --> 01:07:03.300 align:middle line:90% And now look. 01:07:03.300 --> 01:07:05.380 align:middle line:84% The person-wise indifference curve no longer 01:07:05.380 --> 01:07:07.600 align:middle line:84% crosses under that horizontal segment 01:07:07.600 --> 01:07:09.900 align:middle line:84% because they don't want to be on welfare anymore. 01:07:09.900 --> 01:07:12.540 align:middle line:84% So by increasing outside opportunities, 01:07:12.540 --> 01:07:15.380 align:middle line:84% we've actually lowered our redistribution spending. 01:07:15.380 --> 01:07:17.300 align:middle line:90% It's sort of simple, right? 01:07:17.300 --> 01:07:19.780 align:middle line:84% Make people well off, so they don't need welfare. 01:07:19.780 --> 01:07:22.420 align:middle line:84% It's a fancy diagram for a pretty simple concept, 01:07:22.420 --> 01:07:25.620 align:middle line:84% which is to make people better off so don't need welfare. 01:07:25.620 --> 01:07:27.780 align:middle line:84% So the question is, how do you do that? 01:07:27.780 --> 01:07:31.180 align:middle line:84% How do you increase people's outside options, 01:07:31.180 --> 01:07:33.800 align:middle line:84% so that they're less reliant on the welfare system? 01:07:33.800 --> 01:07:37.540 align:middle line:84% And there's a number of mechanisms 01:07:37.540 --> 01:07:42.380 align:middle line:84% that we use to increase outside options. 01:07:42.380 --> 01:07:43.940 align:middle line:84% And I'm going to go through them now. 01:07:43.940 --> 01:07:45.000 align:middle line:84% I might not get through them today. 01:07:45.000 --> 01:07:46.680 align:middle line:84% I'll finish it up next time, if not. 01:07:46.680 --> 01:07:49.940 align:middle line:84% So how do we people increase people's outside options? 01:07:49.940 --> 01:07:57.320 align:middle line:84% The first way we do, the classic way, is through training. 01:07:57.320 --> 01:08:03.440 align:middle line:84% That is, let's say, we say to you, look, if you're on welfare, 01:08:03.440 --> 01:08:06.340 align:middle line:84% you have to enter a training program, 01:08:06.340 --> 01:08:08.520 align:middle line:84% so that you can increase your skills 01:08:08.520 --> 01:08:12.080 align:middle line:84% and basically work your way off welfare. 01:08:12.080 --> 01:08:16.560 align:middle line:84% So that's the classic group we called the New Democrats 01:08:16.560 --> 01:08:19.160 align:middle line:84% in the '80s and '90s, which are the Bill Clinton, 01:08:19.160 --> 01:08:22.560 align:middle line:84% more centrist wing of the Democratic Party that took over. 01:08:22.560 --> 01:08:26.960 align:middle line:84% And their view was, we want to help poor people, but we really 01:08:26.960 --> 01:08:30.055 align:middle line:84% we understand economics, and we think moral hazard is a problem. 01:08:30.055 --> 01:08:31.680 align:middle line:84% So what we're going to do is try to get 01:08:31.680 --> 01:08:35.038 align:middle line:84% people to work their way up through training them. 01:08:35.038 --> 01:08:36.830 align:middle line:84% There's been an enormous amount of studies. 01:08:36.830 --> 01:08:39.497 align:middle line:84% They actually ran a huge amount of randomized trials of training 01:08:39.497 --> 01:08:41.279 align:middle line:90% programs across states. 01:08:41.279 --> 01:08:43.520 align:middle line:90% And the evidence is meh. 01:08:43.520 --> 01:08:46.880 align:middle line:90% Basically, training works a bit. 01:08:46.880 --> 01:08:49.620 align:middle line:84% Essentially, if you train people, you get a few of them 01:08:49.620 --> 01:08:50.899 align:middle line:90% off welfare. 01:08:50.899 --> 01:08:52.740 align:middle line:84% It roughly pays for itself in terms 01:08:52.740 --> 01:08:56.146 align:middle line:84% of how much welfare spending you get but it's not transformative. 01:08:56.146 --> 01:08:58.479 align:middle line:84% It doesn't really get a whole lot of people off welfare. 01:08:58.479 --> 01:09:02.140 align:middle line:84% It kind of works, but not incredibly well. 01:09:02.140 --> 01:09:04.319 align:middle line:90% So that was the first approach. 01:09:04.319 --> 01:09:04.819 align:middle line:90% Yeah? 01:09:04.819 --> 01:09:07.039 align:middle line:90% AUDIENCE: [INAUDIBLE] 01:09:07.039 --> 01:09:10.060 align:middle line:84% PROFESSOR: It's paid from the government budget. 01:09:10.060 --> 01:09:11.819 align:middle line:90% That's the first approach. 01:09:11.819 --> 01:09:17.060 align:middle line:84% The second approach is, let's raise the minimum wage. 01:09:17.060 --> 01:09:20.000 align:middle line:84% If people earn more in the labor market, they won't need welfare. 01:09:20.000 --> 01:09:20.939 align:middle line:84% And in fact, there are studies that 01:09:20.939 --> 01:09:22.479 align:middle line:84% show that with a higher minimum wage, 01:09:22.479 --> 01:09:24.396 align:middle line:84% fewer people are getting government transfers. 01:09:24.396 --> 01:09:28.180 align:middle line:84% It's mechanical, if people are earning more. 01:09:28.180 --> 01:09:31.175 align:middle line:84% Now the trick, of course, is it's not mechanical. 01:09:31.175 --> 01:09:31.800 align:middle line:90% Let me back up. 01:09:31.800 --> 01:09:33.740 align:middle line:84% It's not mechanical because we have the standard problem 01:09:33.740 --> 01:09:34.657 align:middle line:90% with the minimum wage. 01:09:34.657 --> 01:09:36.620 align:middle line:84% This isn't a labor economics class. 01:09:36.620 --> 01:09:39.984 align:middle line:84% How many people took 1403 with David Autor? 01:09:39.984 --> 01:09:41.359 align:middle line:84% Did you take it with David Autor? 01:09:41.359 --> 01:09:42.399 align:middle line:90% So he talks a lot about this. 01:09:42.399 --> 01:09:43.399 align:middle line:90% He's done a lot of work. 01:09:43.399 --> 01:09:44.498 align:middle line:90% I mentioned it in 14.01. 01:09:44.498 --> 01:09:46.540 align:middle line:84% The bottom line, the question of the minimum wage 01:09:46.540 --> 01:09:50.372 align:middle line:84% in a standard economic model, is by raising the wage floor, 01:09:50.372 --> 01:09:52.080 align:middle line:84% you interfere with the free labor market, 01:09:52.080 --> 01:09:54.720 align:middle line:84% and you could cause people to lose their jobs. 01:09:54.720 --> 01:09:56.840 align:middle line:84% But in fact, there's now been decades of evidence 01:09:56.840 --> 01:09:59.400 align:middle line:84% which suggest that's not really true, that, in fact, we 01:09:59.400 --> 01:10:02.280 align:middle line:84% live in a world of imperfect labor markets 01:10:02.280 --> 01:10:05.960 align:middle line:84% where employers have monopsony power 01:10:05.960 --> 01:10:07.793 align:middle line:84% and therefore, in that world, you 01:10:07.793 --> 01:10:09.960 align:middle line:84% can actually raise the minimum wage without creating 01:10:09.960 --> 01:10:11.600 align:middle line:90% unemployment. 01:10:11.600 --> 01:10:14.420 align:middle line:84% And at least in the range where we've changed the minimum wage, 01:10:14.420 --> 01:10:16.080 align:middle line:90% that seems to be largely true. 01:10:16.080 --> 01:10:17.622 align:middle line:84% I think the consensus of the evidence 01:10:17.622 --> 01:10:20.880 align:middle line:84% is in the range of $5 to $10 an hour, 01:10:20.880 --> 01:10:22.755 align:middle line:90% we've changed the minimum wage. 01:10:22.755 --> 01:10:24.880 align:middle line:84% It looks like changing minimum wage does not really 01:10:24.880 --> 01:10:26.240 align:middle line:90% cause unemployment. 01:10:26.240 --> 01:10:32.160 align:middle line:84% Of course, now the minimum wage is much higher than it was. 01:10:32.160 --> 01:10:34.220 align:middle line:84% It's now going to about $15 an hour. 01:10:34.220 --> 01:10:37.080 align:middle line:84% The evidence that level is a little more mixed. 01:10:37.080 --> 01:10:39.780 align:middle line:84% So there is some point at which even the best 01:10:39.780 --> 01:10:41.280 align:middle line:84% evidence of raising the minimum wage 01:10:41.280 --> 01:10:44.280 align:middle line:84% would agree that there is some point at which raising 01:10:44.280 --> 01:10:47.500 align:middle line:84% the minimum wage will cause unemployment. 01:10:47.500 --> 01:10:50.760 align:middle line:84% It's clearly a point above $10 an hour. 01:10:50.760 --> 01:10:54.660 align:middle line:84% And remember, our federal minimum wage is $7.25 an hour. 01:10:54.660 --> 01:10:57.705 align:middle line:84% So it's clearly higher than where the minimum wage is today. 01:10:57.705 --> 01:10:59.580 align:middle line:84% Whether it's as high as where minimum wage is 01:10:59.580 --> 01:11:03.940 align:middle line:84% in places like Massachusetts, where it's $15 an hour, I think, 01:11:03.940 --> 01:11:05.980 align:middle line:84% likely, the answer is yes, that you can do that 01:11:05.980 --> 01:11:07.440 align:middle line:90% without causing unemployment. 01:11:07.440 --> 01:11:10.780 align:middle line:84% But I'd say the evidence is still a little more mixed 01:11:10.780 --> 01:11:12.220 align:middle line:90% on that. 01:11:12.220 --> 01:11:13.340 align:middle line:90% Enoch you have a quick? 01:11:13.340 --> 01:11:14.100 align:middle line:84% AUDIENCE: Yeah, I was going to ask, 01:11:14.100 --> 01:11:15.725 align:middle line:84% because I know the Congressional Budget 01:11:15.725 --> 01:11:19.300 align:middle line:84% Office, when that was being discussed, did a lot of research 01:11:19.300 --> 01:11:20.000 align:middle line:90% on that. 01:11:20.000 --> 01:11:22.180 align:middle line:84% And I think they even have an interactive simulation 01:11:22.180 --> 01:11:24.460 align:middle line:84% on how much you raise the minimum wage and-- 01:11:24.460 --> 01:11:26.720 align:middle line:84% PROFESSOR: The displacement effects. 01:11:26.720 --> 01:11:28.660 align:middle line:84% Yeah, but that's their estimates, 01:11:28.660 --> 01:11:30.880 align:middle line:84% and they do as well as they can with that. 01:11:30.880 --> 01:11:33.500 align:middle line:84% But I think the bottom line is, you can only 01:11:33.500 --> 01:11:36.195 align:middle line:84% live with the weight the evidence is. 01:11:36.195 --> 01:11:38.320 align:middle line:84% We don't have that many experiments of $15 an hour, 01:11:38.320 --> 01:11:39.240 align:middle line:90% so you don't care. 01:11:39.240 --> 01:11:41.160 align:middle line:84% And we think we think it could matter. 01:11:41.160 --> 01:11:42.420 align:middle line:90% This isn't linear. 01:11:42.420 --> 01:11:45.185 align:middle line:84% Clearly, a raise the minimum wage from $1 to $2 01:11:45.185 --> 01:11:47.310 align:middle line:84% will have no effect on employment because everybody 01:11:47.310 --> 01:11:48.830 align:middle line:90% makes more than $2. 01:11:48.830 --> 01:11:50.532 align:middle line:84% From raising the minimum wage to $100, 01:11:50.532 --> 01:11:51.990 align:middle line:84% it would cause massive unemployment 01:11:51.990 --> 01:11:54.050 align:middle line:84% because most people don't make $100 an hour. 01:11:54.050 --> 01:11:56.228 align:middle line:84% So the question is how that function looks. 01:11:56.228 --> 01:11:58.270 align:middle line:84% And there may be estimates, but they're still not 01:11:58.270 --> 01:12:00.710 align:middle line:90% really well identified. 01:12:00.710 --> 01:12:03.070 align:middle line:84% So we could raise the minimum wage, 01:12:03.070 --> 01:12:05.418 align:middle line:84% but we certainly know whether or not 01:12:05.418 --> 01:12:07.710 align:middle line:84% the cause of unemployment-- there's clear evidence that 01:12:07.710 --> 01:12:10.252 align:middle line:84% raising the minimum wage leads to less spending on government 01:12:10.252 --> 01:12:10.930 align:middle line:90% transfers. 01:12:10.930 --> 01:12:14.670 align:middle line:90% 01:12:14.670 --> 01:12:16.950 align:middle line:84% There's a study that finds that for every $1 increase 01:12:16.950 --> 01:12:22.370 align:middle line:84% in the minimum wage, government expenditures decrease by 4%. 01:12:22.370 --> 01:12:24.662 align:middle line:90% 01:12:24.662 --> 01:12:26.870 align:middle line:84% Clearly, another way to reduce government expenditure 01:12:26.870 --> 01:12:28.078 align:middle line:90% is to raise the minimum wage. 01:12:28.078 --> 01:12:32.953 align:middle line:84% That's another way to increase outside options 01:12:32.953 --> 01:12:34.870 align:middle line:84% with the potential controversy over whether it 01:12:34.870 --> 01:12:37.430 align:middle line:90% causes unemployment. 01:12:37.430 --> 01:12:41.210 align:middle line:84% A third approach that's a little more subtle and interesting 01:12:41.210 --> 01:12:50.850 align:middle line:84% is labor market subsidies, which is what if you actually 01:12:50.850 --> 01:12:53.130 align:middle line:90% paid people to work? 01:12:53.130 --> 01:12:55.570 align:middle line:84% What if instead of tax equals work, 01:12:55.570 --> 01:12:57.730 align:middle line:90% you actually paid them to work? 01:12:57.730 --> 01:12:59.930 align:middle line:84% So for example-- we'll talk about this much more 01:12:59.930 --> 01:13:01.730 align:middle line:84% in chapter 21-- we have a program in the US 01:13:01.730 --> 01:13:03.930 align:middle line:84% called the earned income tax credit. 01:13:03.930 --> 01:13:05.670 align:middle line:84% And we'll talk a lot more in chapter 21, 01:13:05.670 --> 01:13:07.270 align:middle line:84% so I'm just going to go briefly through that here. 01:13:07.270 --> 01:13:09.687 align:middle line:84% It's actually the government's largest cash redistribution 01:13:09.687 --> 01:13:11.050 align:middle line:90% program. 01:13:11.050 --> 01:13:14.890 align:middle line:84% The way the EITC works is, for people who are poor, 01:13:14.890 --> 01:13:17.070 align:middle line:84% for every dollar you earn, instead of paying taxes, 01:13:17.070 --> 01:13:19.290 align:middle line:84% you get $0.40 from the government. 01:13:19.290 --> 01:13:21.370 align:middle line:90% So it's a subsidy to work. 01:13:21.370 --> 01:13:24.497 align:middle line:84% For every dollar you earn, you actually get $0.40 more from 01:13:24.497 --> 01:13:26.330 align:middle line:84% the government rather than paying tax on it, 01:13:26.330 --> 01:13:27.310 align:middle line:90% up to a certain level. 01:13:27.310 --> 01:13:29.410 align:middle line:90% Then you start to pay tax again. 01:13:29.410 --> 01:13:33.850 align:middle line:84% And the idea is that basically, by doing 01:13:33.850 --> 01:13:39.130 align:middle line:84% that, you will get people to go to work and get off welfare. 01:13:39.130 --> 01:13:42.590 align:middle line:84% The problem is that it's incredibly expensive. 01:13:42.590 --> 01:13:45.870 align:middle line:84% And if you think about how we spend $15 billion a year 01:13:45.870 --> 01:13:48.670 align:middle line:84% on cash welfare and the EITC is $80 billion a year, 01:13:48.670 --> 01:13:50.670 align:middle line:84% that can't really be a good enough justification 01:13:50.670 --> 01:13:52.430 align:middle line:90% for the EITC. 01:13:52.430 --> 01:13:55.190 align:middle line:84% And that's why other people have suggested a much more targeted 01:13:55.190 --> 01:14:00.030 align:middle line:84% form of this, which is take people who are on welfare today 01:14:00.030 --> 01:14:02.353 align:middle line:90% and pay them to get off. 01:14:02.353 --> 01:14:04.270 align:middle line:84% So basically, the best evidence is the program 01:14:04.270 --> 01:14:08.430 align:middle line:84% in Canada, which was the SSP-- what does that stand for, SSP? 01:14:08.430 --> 01:14:12.230 align:middle line:90% Self-Sufficiency Project. 01:14:12.230 --> 01:14:14.870 align:middle line:84% It was a random experiment, and they 01:14:14.870 --> 01:14:18.870 align:middle line:84% went to Canadian welfare recipients who'd been receiving 01:14:18.870 --> 01:14:20.790 align:middle line:90% welfare for more than one year. 01:14:20.790 --> 01:14:24.830 align:middle line:84% And they said, if you find a full-time job, 01:14:24.830 --> 01:14:30.750 align:middle line:84% we will, for every dollar you earn, give you another dollar. 01:14:30.750 --> 01:14:32.582 align:middle line:84% We'll literally double your earnings. 01:14:32.582 --> 01:14:34.790 align:middle line:84% And they randomly said that to some people on welfare 01:14:34.790 --> 01:14:36.830 align:middle line:84% and to some people, they didn't say that. 01:14:36.830 --> 01:14:40.330 align:middle line:84% And what they found was an enormous response. 01:14:40.330 --> 01:14:43.090 align:middle line:84% They found that basically, there was 01:14:43.090 --> 01:14:46.770 align:middle line:84% almost a 50% increase in the employment rate of the treatment 01:14:46.770 --> 01:14:47.378 align:middle line:90% group. 01:14:47.378 --> 01:14:49.170 align:middle line:84% There was a huge increase in the employment 01:14:49.170 --> 01:14:49.990 align:middle line:90% rate of treatment group. 01:14:49.990 --> 01:14:51.890 align:middle line:84% And in fact, they increased their earnings 01:14:51.890 --> 01:14:54.970 align:middle line:84% so much that the program paid for itself. 01:14:54.970 --> 01:14:57.650 align:middle line:84% If you looked at how much less the government spent on welfare 01:14:57.650 --> 01:15:00.590 align:middle line:84% and how much the government ultimately collected in taxes, 01:15:00.590 --> 01:15:02.970 align:middle line:90% the program paid for itself. 01:15:02.970 --> 01:15:05.910 align:middle line:84% The problem was, once the program ended, 01:15:05.910 --> 01:15:07.850 align:middle line:90% the effect went away. 01:15:07.850 --> 01:15:10.410 align:middle line:84% So the hope was that you'd get people used to working, 01:15:10.410 --> 01:15:11.690 align:middle line:84% and then you could pull the program away 01:15:11.690 --> 01:15:12.390 align:middle line:90% and it would still work. 01:15:12.390 --> 01:15:12.890 align:middle line:90% It didn't. 01:15:12.890 --> 01:15:14.038 align:middle line:90% That didn't happen. 01:15:14.038 --> 01:15:15.830 align:middle line:84% So it looks like to make this program work, 01:15:15.830 --> 01:15:17.038 align:middle line:90% we've just got to keep going. 01:15:17.038 --> 01:15:21.930 align:middle line:90% 01:15:21.930 --> 01:15:25.570 align:middle line:84% So basically, that leads to the really interesting trade 01:15:25.570 --> 01:15:32.970 align:middle line:84% off-between general wage subsidies and targeted wage 01:15:32.970 --> 01:15:35.860 align:middle line:84% subsidies, which is that the general wage 01:15:35.860 --> 01:15:39.100 align:middle line:90% subsidy is way more expensive. 01:15:39.100 --> 01:15:41.340 align:middle line:90% That's the disadvantage. 01:15:41.340 --> 01:15:44.980 align:middle line:84% The target is cheaper, but what about the person 01:15:44.980 --> 01:15:46.487 align:middle line:90% who didn't get welfare? 01:15:46.487 --> 01:15:48.820 align:middle line:84% What if I'm just a person who felt like I shouldn't take 01:15:48.820 --> 01:15:50.120 align:middle line:90% a handout from the government? 01:15:50.120 --> 01:15:51.640 align:middle line:90% Then I don't benefit from this. 01:15:51.640 --> 01:15:53.580 align:middle line:90% That seems a bit unfair. 01:15:53.580 --> 01:15:55.508 align:middle line:84% So let's say Paul and I are equally poor. 01:15:55.508 --> 01:15:57.300 align:middle line:84% I'm someone who's not embarrassed by taking 01:15:57.300 --> 01:15:58.280 align:middle line:90% a check from the government. 01:15:58.280 --> 01:15:58.920 align:middle line:90% He is. 01:15:58.920 --> 01:16:01.860 align:middle line:84% Then, suddenly, I get this huge benefit and he doesn't? 01:16:01.860 --> 01:16:03.380 align:middle line:90% That seems kind of unfair. 01:16:03.380 --> 01:16:07.460 align:middle line:84% Moreover, what pernicious effect could it have? 01:16:07.460 --> 01:16:09.820 align:middle line:84% What could be the pernicious effect if I said that once 01:16:09.820 --> 01:16:12.237 align:middle line:84% you're on welfare, you get this program and if you're not, 01:16:12.237 --> 01:16:13.060 align:middle line:90% you don't. 01:16:13.060 --> 01:16:14.357 align:middle line:90% Yeah? 01:16:14.357 --> 01:16:15.440 align:middle line:90% People could join welfare. 01:16:15.440 --> 01:16:19.540 align:middle line:84% It's what's called the entry effect. 01:16:19.540 --> 01:16:21.640 align:middle line:84% Basically, you could actually end up 01:16:21.640 --> 01:16:23.140 align:middle line:84% creating a whole new class of people 01:16:23.140 --> 01:16:26.900 align:middle line:84% on welfare because they joined to qualify for the program. 01:16:26.900 --> 01:16:29.860 align:middle line:84% Now, it turned out with the SSP, that didn't happen. 01:16:29.860 --> 01:16:33.100 align:middle line:84% Because with the SSP, you had to be on welfare for a whole year 01:16:33.100 --> 01:16:35.213 align:middle line:84% to qualify, and there weren't many people 01:16:35.213 --> 01:16:37.380 align:middle line:84% who wanted to wait a whole year to get this subsidy. 01:16:37.380 --> 01:16:38.922 align:middle line:84% So even though there's a big subsidy, 01:16:38.922 --> 01:16:41.720 align:middle line:84% they weren't willing to go on welfare for a year to get it. 01:16:41.720 --> 01:16:44.120 align:middle line:84% But by making them Wait a year, you 01:16:44.120 --> 01:16:45.905 align:middle line:84% get lots of people who deserved it, 01:16:45.905 --> 01:16:48.280 align:middle line:84% who maybe dropped out after six months and didn't get it. 01:16:48.280 --> 01:16:49.640 align:middle line:90% So that's the trade-off. 01:16:49.640 --> 01:16:51.680 align:middle line:84% The more you target it, the more targeted 01:16:51.680 --> 01:16:53.060 align:middle line:90% it is, the less money you spend. 01:16:53.060 --> 01:16:54.727 align:middle line:84% But there are people you're not helping. 01:16:54.727 --> 01:16:56.320 align:middle line:90% The ITC helps everyone. 01:16:56.320 --> 01:16:58.220 align:middle line:84% Everyone gets low-income benefits from it, 01:16:58.220 --> 01:17:00.600 align:middle line:90% but it costs a ton more money. 01:17:00.600 --> 01:17:03.000 align:middle line:84% That's why we have social welfare functions. 01:17:03.000 --> 01:17:06.410 align:middle line:84% As we did in 14.01, essentially, you want to ask yourself, 01:17:06.410 --> 01:17:08.160 align:middle line:84% what is the benefit of this bigger program 01:17:08.160 --> 01:17:12.020 align:middle line:84% in terms of the moral hazard it costs and the benefit it gives? 01:17:12.020 --> 01:17:14.380 align:middle line:84% It turns out-- we'll show you in chapter 21-- 01:17:14.380 --> 01:17:17.480 align:middle line:84% there is no trade-off for the EITC The EITC actually 01:17:17.480 --> 01:17:22.800 align:middle line:84% is a patch to the bucket, that it redistributes and makes 01:17:22.800 --> 01:17:24.240 align:middle line:90% money for the government. 01:17:24.240 --> 01:17:27.620 align:middle line:84% And we'll talk about that next time or chapter 21. 01:17:27.620 --> 01:17:29.630 align:middle line:90% Let's stop there. 01:17:29.630 --> 01:17:33.000 align:middle line:90%