WEBVTT 00:00:00.000 --> 00:00:01.937 align:middle line:90% [SQUEAKING] 00:00:01.937 --> 00:00:03.853 align:middle line:90% [RUSTLING] 00:00:03.853 --> 00:00:07.206 align:middle line:90% [CLICKING] 00:00:07.206 --> 00:00:10.080 align:middle line:90% 00:00:10.080 --> 00:00:12.480 align:middle line:84% JON GRUBER: Today, we're going to continue 00:00:12.480 --> 00:00:14.520 align:middle line:90% with discussion of chapter 7. 00:00:14.520 --> 00:00:19.920 align:middle line:84% So you remember chapter 7, we started with basically defining 00:00:19.920 --> 00:00:22.640 align:middle line:84% public goods, which are goods that are 00:00:22.640 --> 00:00:25.400 align:middle line:90% non-rival and non-excludable. 00:00:25.400 --> 00:00:32.400 align:middle line:84% We then talked about how basically, the optimal provision 00:00:32.400 --> 00:00:34.680 align:middle line:84% of public goods would be determined differently 00:00:34.680 --> 00:00:36.880 align:middle line:84% than optimal provision of private goods. 00:00:36.880 --> 00:00:39.700 align:middle line:84% Then when you solve the optimal provision of private goods, 00:00:39.700 --> 00:00:41.720 align:middle line:84% you're equating each individual's marginal rate 00:00:41.720 --> 00:00:45.360 align:middle line:84% of substitution with the marginal rate of transformation 00:00:45.360 --> 00:00:48.520 align:middle line:84% or in this case, just marginal cost. 00:00:48.520 --> 00:00:49.920 align:middle line:84% But when you do public goods, you 00:00:49.920 --> 00:00:54.510 align:middle line:84% care about the sum of marginal rates of substitution, 00:00:54.510 --> 00:00:56.760 align:middle line:84% because when you have a public good, everyone gets it. 00:00:56.760 --> 00:00:58.385 align:middle line:84% So you care about everyone's happiness, 00:00:58.385 --> 00:01:01.070 align:middle line:90% not just one person's happiness. 00:01:01.070 --> 00:01:04.790 align:middle line:84% So that's the key difference between the two. 00:01:04.790 --> 00:01:08.030 align:middle line:84% And that key difference drives the answer 00:01:08.030 --> 00:01:10.590 align:middle line:84% to the first question in public finance, which is when should 00:01:10.590 --> 00:01:11.632 align:middle line:90% the government intervene. 00:01:11.632 --> 00:01:14.590 align:middle line:84% And the answer is that the government needs to intervene. 00:01:14.590 --> 00:01:15.970 align:middle line:90% Or let me say it again. 00:01:15.970 --> 00:01:19.870 align:middle line:84% The market will fail if there is-- 00:01:19.870 --> 00:01:23.010 align:middle line:84% the private market will fail to provide other goods, 00:01:23.010 --> 00:01:25.790 align:middle line:84% because the classic free rider problem. 00:01:25.790 --> 00:01:28.510 align:middle line:84% Reminder-- the free rider problem is whenever 00:01:28.510 --> 00:01:34.070 align:middle line:84% someone's actions have a private cost, but a public benefit, 00:01:34.070 --> 00:01:37.510 align:middle line:84% they will underprovide that action 00:01:37.510 --> 00:01:40.070 align:middle line:84% because they will not see all the benefit. 00:01:40.070 --> 00:01:44.670 align:middle line:84% So to see that, I have an example in the text, but let's 00:01:44.670 --> 00:01:47.150 align:middle line:84% go to the example in the appendix, because the appendix 00:01:47.150 --> 00:01:48.610 align:middle line:90% here is for MIT students. 00:01:48.610 --> 00:01:51.070 align:middle line:90% The text is for others-- 00:01:51.070 --> 00:01:52.130 align:middle line:90% MIT students too. 00:01:52.130 --> 00:01:56.230 align:middle line:84% But the appendix is more what you want to see and can handle. 00:01:56.230 --> 00:01:59.130 align:middle line:84% So let's take two individuals, Ben and Jerry. 00:01:59.130 --> 00:02:01.170 align:middle line:84% Ben and Jerry live by themselves, far away 00:02:01.170 --> 00:02:02.650 align:middle line:90% from others. 00:02:02.650 --> 00:02:06.770 align:middle line:84% They have a private good x with a price of $1. 00:02:06.770 --> 00:02:08.270 align:middle line:90% So P sub x equals 1. 00:02:08.270 --> 00:02:09.729 align:middle line:90% That's the numerator good. 00:02:09.729 --> 00:02:13.090 align:middle line:84% They consume two things-- the private good, x with P 00:02:13.090 --> 00:02:17.570 align:middle line:84% sub x is 1 and fireworks which is a pure public good. 00:02:17.570 --> 00:02:21.690 align:middle line:84% And we're also going to assume P sub f equals 1. 00:02:21.690 --> 00:02:23.510 align:middle line:84% Once again, no harm in doing this. 00:02:23.510 --> 00:02:26.610 align:middle line:84% It just avoids us carrying around a bunch of extra P terms. 00:02:26.610 --> 00:02:28.390 align:middle line:84% It doesn't change the math at all. 00:02:28.390 --> 00:02:30.210 align:middle line:90% It just makes it simpler to see. 00:02:30.210 --> 00:02:33.490 align:middle line:84% So they can consume a private good x. 00:02:33.490 --> 00:02:36.170 align:middle line:84% Think of this as a composite of everything else they consume-- 00:02:36.170 --> 00:02:39.290 align:middle line:84% or fireworks, which is a pure public good. 00:02:39.290 --> 00:02:43.850 align:middle line:84% They each have an income of $100. 00:02:43.850 --> 00:02:48.690 align:middle line:90% So yB equals yj equals 100. 00:02:48.690 --> 00:02:51.930 align:middle line:84% They each have an income of $100. 00:02:51.930 --> 00:02:54.530 align:middle line:84% And because fireworks are a public good, 00:02:54.530 --> 00:02:57.220 align:middle line:84% they care about the total amount of fireworks, 00:02:57.220 --> 00:03:00.280 align:middle line:84% which is fireworks provided by Ben plus fireworks provided 00:03:00.280 --> 00:03:01.880 align:middle line:90% by Jerry. 00:03:01.880 --> 00:03:04.840 align:middle line:84% What they care about is the total amount of fireworks. 00:03:04.840 --> 00:03:07.920 align:middle line:84% In particular, each individual i has 00:03:07.920 --> 00:03:09.780 align:middle line:90% utility of the following form. 00:03:09.780 --> 00:03:12.460 align:middle line:84% U-- we're going to assume their utility. 00:03:12.460 --> 00:03:14.680 align:middle line:84% Obviously, utility functions are assumed. 00:03:14.680 --> 00:03:18.320 align:middle line:84% Their utility function is 2 times 00:03:18.320 --> 00:03:24.520 align:middle line:84% log x sub i where I denotes b and j-- that's 00:03:24.520 --> 00:03:31.520 align:middle line:84% each individual, plus log f, which is 00:03:31.520 --> 00:03:32.980 align:middle line:90% the total amount of fireworks. 00:03:32.980 --> 00:03:34.100 align:middle line:90% Here's the key thing. 00:03:34.100 --> 00:03:36.100 align:middle line:84% This is what's going to make the math work here, 00:03:36.100 --> 00:03:38.720 align:middle line:84% which is that they only care about the total amount 00:03:38.720 --> 00:03:39.340 align:middle line:90% of fireworks. 00:03:39.340 --> 00:03:41.340 align:middle line:84% They don't care about the amount they set off. 00:03:41.340 --> 00:03:43.640 align:middle line:90% It's just what's in the sky. 00:03:43.640 --> 00:03:45.560 align:middle line:84% We'll come back later to when they get utility 00:03:45.560 --> 00:03:47.760 align:middle line:84% from actually setting it off, but we're 00:03:47.760 --> 00:03:49.020 align:middle line:90% going to ignore that for now. 00:03:49.020 --> 00:03:50.600 align:middle line:84% We're just going to assume they're indifferent who 00:03:50.600 --> 00:03:51.558 align:middle line:90% sets the fireworks off. 00:03:51.558 --> 00:03:54.120 align:middle line:84% They just care how many fireworks are in the sky. 00:03:54.120 --> 00:03:57.630 align:middle line:84% And their budget constraint is that x 00:03:57.630 --> 00:04:02.990 align:middle line:90% sub i plus f sub i equals 100. 00:04:02.990 --> 00:04:04.890 align:middle line:84% So that's their maximization problem. 00:04:04.890 --> 00:04:07.550 align:middle line:90% 00:04:07.550 --> 00:04:10.910 align:middle line:84% And the reason it's just quantities rather than amounts 00:04:10.910 --> 00:04:12.717 align:middle line:84% is because we assume prices are equal to 1, 00:04:12.717 --> 00:04:14.550 align:middle line:84% so our budget constraint, which is expressed 00:04:14.550 --> 00:04:16.310 align:middle line:90% in terms of quantities. 00:04:16.310 --> 00:04:21.870 align:middle line:84% Now, imagine that each individual 00:04:21.870 --> 00:04:25.310 align:middle line:84% is deciding on their own how many fireworks to set up. 00:04:25.310 --> 00:04:27.430 align:middle line:84% Now, here we go back to a classic game 00:04:27.430 --> 00:04:29.830 align:middle line:84% theoretic prisoner's dilemma type problem 00:04:29.830 --> 00:04:33.670 align:middle line:84% of the type we saw in 14.01, which is that basically, we 00:04:33.670 --> 00:04:38.110 align:middle line:84% have an issue where each individual's actions affect 00:04:38.110 --> 00:04:39.950 align:middle line:90% the other. 00:04:39.950 --> 00:04:44.350 align:middle line:84% So they're going to have to have some game theoretic solution 00:04:44.350 --> 00:04:45.250 align:middle line:90% to the situation. 00:04:45.250 --> 00:04:49.470 align:middle line:84% And we're going to, as in 14.01, assume that there's 00:04:49.470 --> 00:04:52.070 align:middle line:90% a Nash bargaining context. 00:04:52.070 --> 00:04:55.250 align:middle line:84% And remember, the way the Nash bargaining context works 00:04:55.250 --> 00:04:59.290 align:middle line:84% is that each individual optimizes their behavior, 00:04:59.290 --> 00:05:03.010 align:middle line:84% taking the other individual's behavior as given. 00:05:03.010 --> 00:05:06.050 align:middle line:84% And then a Nash equilibrium is one 00:05:06.050 --> 00:05:08.370 align:middle line:84% where both are satisfied where, when 00:05:08.370 --> 00:05:11.130 align:middle line:84% I do what's best for me, given your behavior, 00:05:11.130 --> 00:05:14.050 align:middle line:84% and you do what's best for you, given my behavior, 00:05:14.050 --> 00:05:16.570 align:middle line:84% we get to the same equilibrium point. 00:05:16.570 --> 00:05:19.250 align:middle line:84% Then we're at a Nash equilibrium. 00:05:19.250 --> 00:05:21.850 align:middle line:84% And so it's just going to be the same kind of math 00:05:21.850 --> 00:05:24.850 align:middle line:84% we did when we did Nash equilibrium, Cournot equilibria 00:05:24.850 --> 00:05:29.010 align:middle line:90% in 14.01. 00:05:29.010 --> 00:05:31.490 align:middle line:84% Basically, we're going to simply say-- 00:05:31.490 --> 00:05:34.970 align:middle line:90% 00:05:34.970 --> 00:05:35.928 align:middle line:90% we can do a trick here. 00:05:35.928 --> 00:05:37.595 align:middle line:84% We can just insert the budget constraint 00:05:37.595 --> 00:05:39.210 align:middle line:84% directly into the utility function 00:05:39.210 --> 00:05:40.810 align:middle line:84% to make our math a little easier. 00:05:40.810 --> 00:05:44.230 align:middle line:84% We can just say their goal is to maximize utility, 00:05:44.230 --> 00:05:49.990 align:middle line:84% which is equal to 2 times log 100 minus fi 00:05:49.990 --> 00:05:59.720 align:middle line:90% plus log fi plus fj. 00:05:59.720 --> 00:06:01.680 align:middle line:84% So I've now written their utility solely 00:06:01.680 --> 00:06:03.000 align:middle line:90% as a function of fireworks. 00:06:03.000 --> 00:06:05.600 align:middle line:90% 00:06:05.600 --> 00:06:07.900 align:middle line:84% So let's make this utility of Jerry. 00:06:07.900 --> 00:06:09.800 align:middle line:90% Let me write this right-- 00:06:09.800 --> 00:06:14.120 align:middle line:90% 00:06:14.120 --> 00:06:19.040 align:middle line:84% utility of Ben, so it's 2 times 100 minus 00:06:19.040 --> 00:06:22.440 align:middle line:84% how many fireworks Ben sets off, plus the fireworks set by Ben 00:06:22.440 --> 00:06:23.900 align:middle line:90% and the fireworks set by Jerry. 00:06:23.900 --> 00:06:26.848 align:middle line:90% 00:06:26.848 --> 00:06:28.140 align:middle line:90% That is their utility function. 00:06:28.140 --> 00:06:30.520 align:middle line:84% Just plug the budget constraint in. 00:06:30.520 --> 00:06:37.160 align:middle line:84% If we differentiate this with respect to f sub B, 00:06:37.160 --> 00:06:47.480 align:middle line:84% you get minus 2 over 100 minus f sub B 00:06:47.480 --> 00:06:56.550 align:middle line:84% plus 1 over f sub B plus f sub j equals 0. 00:06:56.550 --> 00:07:01.030 align:middle line:84% That's what happens when we differentiate that equation. 00:07:01.030 --> 00:07:07.750 align:middle line:84% And therefore from that, we can generate the conclusion-- 00:07:07.750 --> 00:07:09.550 align:middle line:90% just going to work upwards here. 00:07:09.550 --> 00:07:14.470 align:middle line:84% We'll generate the conclusion that f sub B is equal 00:07:14.470 --> 00:07:21.630 align:middle line:90% to 100 minus 2 f sub J over 3. 00:07:21.630 --> 00:07:24.310 align:middle line:90% 00:07:24.310 --> 00:07:27.870 align:middle line:84% F sub B equals 100 minus 2 f sub j over 3. 00:07:27.870 --> 00:07:32.270 align:middle line:84% So once again, taking f sub J is fixed, 00:07:32.270 --> 00:07:34.470 align:middle line:84% Ben solves and gets that outcome. 00:07:34.470 --> 00:07:37.390 align:middle line:84% Similarly, by symmetry, we get that f 00:07:37.390 --> 00:07:45.530 align:middle line:84% sub j equals 100 minus 2 f sub B over 3. 00:07:45.530 --> 00:07:46.610 align:middle line:90% You can solve that out. 00:07:46.610 --> 00:07:48.235 align:middle line:84% Or you can just know by symmetry that's 00:07:48.235 --> 00:07:49.970 align:middle line:90% going to be true in this case. 00:07:49.970 --> 00:07:52.730 align:middle line:84% So that's two equations and two unknowns. 00:07:52.730 --> 00:07:56.770 align:middle line:84% Solve them, and you get that f sub B star, 00:07:56.770 --> 00:07:59.170 align:middle line:84% or f sub B the private equilibrium, is 20 00:07:59.170 --> 00:08:02.810 align:middle line:84% and f sub j equals 20, that they each provide 20 00:08:02.810 --> 00:08:05.150 align:middle line:90% fireworks for a total f of 40. 00:08:05.150 --> 00:08:08.450 align:middle line:90% 00:08:08.450 --> 00:08:11.650 align:middle line:90% Questions about that? 00:08:11.650 --> 00:08:17.530 align:middle line:84% So once again, they've each maximized their firework use, 00:08:17.530 --> 00:08:20.490 align:middle line:84% taking the other person's behavior as given. 00:08:20.490 --> 00:08:26.490 align:middle line:84% Now, what is the optimal level of fireworks? 00:08:26.490 --> 00:08:28.090 align:middle line:84% Well, the optimal level of fireworks 00:08:28.090 --> 00:08:32.530 align:middle line:84% we know from last lecture is where 00:08:32.530 --> 00:08:35.010 align:middle line:84% the sum of the marginal rates of substitution 00:08:35.010 --> 00:08:37.530 align:middle line:84% equals the marginal rate of transformation. 00:08:37.530 --> 00:08:42.610 align:middle line:84% So that is the sum of what's the marginal rate of substitution 00:08:42.610 --> 00:08:46.880 align:middle line:90% of Ben. 00:08:46.880 --> 00:08:49.520 align:middle line:84% It's going to be the it's going to be 00:08:49.520 --> 00:08:52.840 align:middle line:84% the ratio of their marginal utility of fireworks 00:08:52.840 --> 00:08:55.960 align:middle line:84% to their marginal utility of private goods. 00:08:55.960 --> 00:08:58.600 align:middle line:84% So the marginal rate of substitution for Ben 00:08:58.600 --> 00:09:07.040 align:middle line:84% equals the ratio of the marginal utility of fireworks to Ben 00:09:07.040 --> 00:09:10.260 align:middle line:84% over the marginal utility of private goods for Ben. 00:09:10.260 --> 00:09:15.080 align:middle line:90% 00:09:15.080 --> 00:09:19.607 align:middle line:84% So if we differentiate the utility function to get this 00:09:19.607 --> 00:09:20.940 align:middle line:90% we're going to get that for Ben. 00:09:20.940 --> 00:09:26.540 align:middle line:84% We're going to get that 100 minus f sub j. 00:09:26.540 --> 00:09:29.200 align:middle line:90% 00:09:29.200 --> 00:09:36.880 align:middle line:84% I'm sorry, minus f sub B. That's differentiating this utility 00:09:36.880 --> 00:09:48.790 align:middle line:84% function with respect to fireworks over 2 times 00:09:48.790 --> 00:09:53.950 align:middle line:84% f sub B plus f sub j is going to be his marginal rate 00:09:53.950 --> 00:09:55.230 align:middle line:90% of substitution. 00:09:55.230 --> 00:09:58.070 align:middle line:84% That's going to be Ben's rate of marginal rate of substitution. 00:09:58.070 --> 00:10:03.550 align:middle line:84% And what do we know from optimal provision of public goods? 00:10:03.550 --> 00:10:07.150 align:middle line:84% That you want to set the sum of marginal rates of substitution 00:10:07.150 --> 00:10:09.868 align:middle line:84% equal to the marginal rate of transformation. 00:10:09.868 --> 00:10:11.910 align:middle line:84% So when I said that, we want to do the same thing 00:10:11.910 --> 00:10:14.570 align:middle line:84% for Jerry, which by symmetry, now all look the same. 00:10:14.570 --> 00:10:24.390 align:middle line:84% It will be 100 minus f sub j over 2 times fB plus fj. 00:10:24.390 --> 00:10:28.450 align:middle line:84% That's going to be Jerry's marginal rate of substitution. 00:10:28.450 --> 00:10:36.110 align:middle line:84% We want to set that sum equal to the ratio of marginal cost 00:10:36.110 --> 00:10:42.630 align:middle line:84% of fireworks over marginal cost of x, which is equal to 1 00:10:42.630 --> 00:10:45.530 align:middle line:90% by assumption here. 00:10:45.530 --> 00:10:47.290 align:middle line:84% Because we're in a competitive market, 00:10:47.290 --> 00:10:49.090 align:middle line:90% price equals marginal cost. 00:10:49.090 --> 00:10:50.690 align:middle line:90% I did skip one step here. 00:10:50.690 --> 00:10:52.630 align:middle line:84% Notice I wrote that prices were 1 here. 00:10:52.630 --> 00:10:54.068 align:middle line:90% I said marginal costs are 1. 00:10:54.068 --> 00:10:55.110 align:middle line:90% Why can I skip that step? 00:10:55.110 --> 00:10:57.810 align:middle line:84% Because in a competitive market, price equals marginal cost. 00:10:57.810 --> 00:10:59.410 align:middle line:90% OK, sorry, that's a-- 00:10:59.410 --> 00:11:02.530 align:middle line:84% be very clear that when I assume price equals 1, 00:11:02.530 --> 00:11:05.490 align:middle line:84% I also assume marginal cost equals 1 because I'm 00:11:05.490 --> 00:11:07.290 align:middle line:90% in a competitive market. 00:11:07.290 --> 00:11:10.130 align:middle line:84% It's just a step I've skipped here. 00:11:10.130 --> 00:11:15.050 align:middle line:84% So this is the equation we solve for optimal provision 00:11:15.050 --> 00:11:16.170 align:middle line:90% of public goods. 00:11:16.170 --> 00:11:22.110 align:middle line:84% And if you solve that, you get that fB equals fj equals 30. 00:11:22.110 --> 00:11:24.650 align:middle line:90% 00:11:24.650 --> 00:11:29.690 align:middle line:84% The optimal provision is that they each provide 30-- 00:11:29.690 --> 00:11:32.050 align:middle line:90% I'm sorry 33.3, my bad-- 00:11:32.050 --> 00:11:40.450 align:middle line:84% 33.3, so that f equals 66.6 or 66.7666. 00:11:40.450 --> 00:11:44.640 align:middle line:90% That is the optimal provision. 00:11:44.640 --> 00:11:48.540 align:middle line:84% That is much higher than the private provision. 00:11:48.540 --> 00:11:52.240 align:middle line:90% 00:11:52.240 --> 00:11:54.880 align:middle line:84% This is the key of all this math. 00:11:54.880 --> 00:12:00.400 align:middle line:84% The key conclusion is that when individuals privately optimize, 00:12:00.400 --> 00:12:04.600 align:middle line:84% taken the other person's behavior as given, 00:12:04.600 --> 00:12:08.160 align:middle line:84% they will end up providing fewer fireworks in total-- 00:12:08.160 --> 00:12:14.600 align:middle line:84% 40 fireworks, rather than 66.66666 fireworks. 00:12:14.600 --> 00:12:15.890 align:middle line:90% Why? 00:12:15.890 --> 00:12:17.140 align:middle line:90% What's the intuition for that? 00:12:17.140 --> 00:12:19.760 align:middle line:84% Someone raise their hand and tell me. 00:12:19.760 --> 00:12:23.800 align:middle line:84% Why does a private provision yield less 00:12:23.800 --> 00:12:27.480 align:middle line:84% than the optimal amount they should be providing? 00:12:27.480 --> 00:12:33.720 align:middle line:84% Hint-- I've already given the answer in this lecture. 00:12:33.720 --> 00:12:35.120 align:middle line:90% Yeah, Enoch. 00:12:35.120 --> 00:12:37.640 align:middle line:84% AUDIENCE: They both benefit from the other's purchase 00:12:37.640 --> 00:12:38.380 align:middle line:90% of fireworks. 00:12:38.380 --> 00:12:40.300 align:middle line:84% JON GRUBER: So it's an example of what? 00:12:40.300 --> 00:12:43.390 align:middle line:90% 00:12:43.390 --> 00:12:44.370 align:middle line:90% Free rider problem. 00:12:44.370 --> 00:12:46.130 align:middle line:84% Raise your hand-- free rider problem. 00:12:46.130 --> 00:12:49.190 align:middle line:84% It's an example of free rider problem. 00:12:49.190 --> 00:12:51.450 align:middle line:84% Basically, this is the key intuition here. 00:12:51.450 --> 00:12:54.590 align:middle line:84% It's tricky, but we have in this class what I call our light bulb 00:12:54.590 --> 00:12:58.733 align:middle line:84% moments where things are hard until they're suddenly not. 00:12:58.733 --> 00:13:00.150 align:middle line:84% You just got to get the light bulb 00:13:00.150 --> 00:13:03.070 align:middle line:84% moment on this, which is that basically, 00:13:03.070 --> 00:13:08.390 align:middle line:84% when my actions benefit other people, 00:13:08.390 --> 00:13:10.870 align:middle line:84% it's the intuition from positive externalities. 00:13:10.870 --> 00:13:15.470 align:middle line:84% A public good is the ultimate positive externality. 00:13:15.470 --> 00:13:20.710 align:middle line:84% When my actions benefit others, I do too little of it. 00:13:20.710 --> 00:13:21.810 align:middle line:90% This is the perfect case. 00:13:21.810 --> 00:13:24.470 align:middle line:84% I'm doing two little fireworks because my actions 00:13:24.470 --> 00:13:26.870 align:middle line:90% are benefiting my neighbor. 00:13:26.870 --> 00:13:29.630 align:middle line:84% So privately, I'm shooting off too few fireworks 00:13:29.630 --> 00:13:32.710 align:middle line:84% relative to what's socially optimal. 00:13:32.710 --> 00:13:38.270 align:middle line:84% So by the Samuelson condition, which we proved last time-- 00:13:38.270 --> 00:13:41.530 align:middle line:84% which says that the optimal provision of public goods 00:13:41.530 --> 00:13:44.330 align:middle line:84% relies on the sum of marginal rate of substitution, 00:13:44.330 --> 00:13:47.050 align:middle line:84% while the private provision of public goods 00:13:47.050 --> 00:13:48.690 align:middle line:84% is going to rely on each individual's 00:13:48.690 --> 00:13:50.330 align:middle line:84% marginal rate of substitution, we 00:13:50.330 --> 00:13:53.710 align:middle line:84% end up providing too few public goods in the private market. 00:13:53.710 --> 00:13:57.210 align:middle line:84% The private market will fail to provide public goods. 00:13:57.210 --> 00:14:00.590 align:middle line:84% Not fail to provide them-- fail to provide the optimal quantity. 00:14:00.590 --> 00:14:02.202 align:middle line:90% And this will come back. 00:14:02.202 --> 00:14:04.410 align:middle line:84% We have to remember to avoid black and white thinking 00:14:04.410 --> 00:14:05.070 align:middle line:90% in this class. 00:14:05.070 --> 00:14:06.945 align:middle line:84% It's not either the private market completely 00:14:06.945 --> 00:14:08.650 align:middle line:90% fails or completely succeeds. 00:14:08.650 --> 00:14:12.610 align:middle line:84% Failure is defined as non-optimality. 00:14:12.610 --> 00:14:15.383 align:middle line:84% It is not defined as the market being destroyed. 00:14:15.383 --> 00:14:16.550 align:middle line:90% It's defined as not optimal. 00:14:16.550 --> 00:14:17.710 align:middle line:90% The market still works. 00:14:17.710 --> 00:14:18.710 align:middle line:90% We get 40 fireworks. 00:14:18.710 --> 00:14:19.750 align:middle line:90% The sky is lit up. 00:14:19.750 --> 00:14:21.530 align:middle line:84% It's just not as lit up as it would 00:14:21.530 --> 00:14:27.530 align:middle line:84% be if there was a social planner optimizing social welfare. 00:14:27.530 --> 00:14:28.670 align:middle line:90% Questions about that? 00:14:28.670 --> 00:14:31.410 align:middle line:90% 00:14:31.410 --> 00:14:32.010 align:middle line:90% Yeah. 00:14:32.010 --> 00:14:35.290 align:middle line:84% AUDIENCE: Do you mind explaining the math behind the 100 minus f 00:14:35.290 --> 00:14:37.210 align:middle line:90% [INAUDIBLE] of x-- or the m-- 00:14:37.210 --> 00:14:39.080 align:middle line:84% JON GRUBER: Oh, the marginal utility-- 00:14:39.080 --> 00:14:44.920 align:middle line:84% I just differentiated this function with respect 00:14:44.920 --> 00:14:47.500 align:middle line:90% to fireworks, f sub B. 00:14:47.500 --> 00:14:49.880 align:middle line:90% AUDIENCE: [INAUDIBLE] data. 00:14:49.880 --> 00:14:53.380 align:middle line:84% JON GRUBER: Yes, exactly, that utility function. 00:14:53.380 --> 00:14:57.840 align:middle line:90% 00:14:57.840 --> 00:15:00.480 align:middle line:90% Other questions? 00:15:00.480 --> 00:15:05.720 align:middle line:84% So this is the mathematical illustration 00:15:05.720 --> 00:15:07.560 align:middle line:90% of the free rider problem. 00:15:07.560 --> 00:15:09.980 align:middle line:84% In practice, the free rider problem is everywhere. 00:15:09.980 --> 00:15:12.200 align:middle line:84% So they have some fun examples in the book. 00:15:12.200 --> 00:15:16.600 align:middle line:84% For example, the state of Victoria in Australia 00:15:16.600 --> 00:15:23.880 align:middle line:84% used to pay for fire protection by a tax on homeowner insurance 00:15:23.880 --> 00:15:24.380 align:middle line:90% policies. 00:15:24.380 --> 00:15:26.005 align:middle line:84% So you bought your homeowner insurance, 00:15:26.005 --> 00:15:27.240 align:middle line:90% there was an extra tax. 00:15:27.240 --> 00:15:30.160 align:middle line:90% And that paid for fire services. 00:15:30.160 --> 00:15:34.800 align:middle line:84% So that meant if you didn't insure your home, 00:15:34.800 --> 00:15:36.455 align:middle line:84% you didn't pay for fire services. 00:15:36.455 --> 00:15:38.580 align:middle line:84% But the fire company still came and saved your home 00:15:38.580 --> 00:15:40.020 align:middle line:84% because they had to because they had to because other homes 00:15:40.020 --> 00:15:41.180 align:middle line:90% had burned down. 00:15:41.180 --> 00:15:43.660 align:middle line:84% So people stopped buying insurance 00:15:43.660 --> 00:15:46.260 align:middle line:84% because they were like, I'm going to get it anyway. 00:15:46.260 --> 00:15:47.700 align:middle line:84% And so then the state of Victoria 00:15:47.700 --> 00:15:49.900 align:middle line:84% switched to financing it through a property tax, 00:15:49.900 --> 00:15:52.460 align:middle line:90% rather than through insurance. 00:15:52.460 --> 00:15:56.460 align:middle line:84% Do we have any New York Citiers here? 00:15:56.460 --> 00:15:57.980 align:middle line:90% No New York Citiers here. 00:15:57.980 --> 00:16:01.400 align:middle line:84% Any of you guys been the Met in New York City, the museum there? 00:16:01.400 --> 00:16:04.820 align:middle line:84% So the Met in New York City, since 1970, 00:16:04.820 --> 00:16:07.658 align:middle line:84% has had a recommended donation instead of an admissions fee. 00:16:07.658 --> 00:16:09.700 align:middle line:84% You don't have to pay to get into this incredibly 00:16:09.700 --> 00:16:10.320 align:middle line:90% famous museum. 00:16:10.320 --> 00:16:12.260 align:middle line:90% There's a recommended donation. 00:16:12.260 --> 00:16:16.080 align:middle line:90% For years, to try to get-- 00:16:16.080 --> 00:16:17.980 align:middle line:84% now, that's a clear free rider problem. 00:16:17.980 --> 00:16:20.440 align:middle line:84% I cannot pay and get the same thing as you get by paying. 00:16:20.440 --> 00:16:22.000 align:middle line:90% So how did they try to solve it? 00:16:22.000 --> 00:16:26.160 align:middle line:84% "Recommended" was in tiny letters relative to the sign. 00:16:26.160 --> 00:16:36.622 align:middle line:84% Well, actually a lawsuit in 2013 said 00:16:36.622 --> 00:16:37.830 align:middle line:90% they could no longer do that. 00:16:37.830 --> 00:16:39.490 align:middle line:84% They had to make it the same size, 00:16:39.490 --> 00:16:44.195 align:middle line:84% and the share of people donating fell from 50% to 10%. 00:16:44.195 --> 00:16:46.070 align:middle line:84% That's another example of free rider problem. 00:16:46.070 --> 00:16:48.278 align:middle line:84% And of course, we all probably, every day, 00:16:48.278 --> 00:16:50.070 align:middle line:84% see examples of this with our online usage. 00:16:50.070 --> 00:16:54.990 align:middle line:84% Take Dropbox, which you can use for free until a certain level. 00:16:54.990 --> 00:16:56.690 align:middle line:90% People don't pay. 00:16:56.690 --> 00:17:00.117 align:middle line:84% Streaming services-- can there's certain options. 00:17:00.117 --> 00:17:00.950 align:middle line:90% And what do they do? 00:17:00.950 --> 00:17:01.825 align:middle line:90% They try to fight it. 00:17:01.825 --> 00:17:02.670 align:middle line:90% Spotify is free. 00:17:02.670 --> 00:17:04.329 align:middle line:84% If it's free, you got to pay ads. 00:17:04.329 --> 00:17:06.069 align:middle line:84% Dropbox, it's better if it's not free, 00:17:06.069 --> 00:17:08.569 align:middle line:84% et cetera That's how they're trying to fight this free rider 00:17:08.569 --> 00:17:09.530 align:middle line:90% problem. 00:17:09.530 --> 00:17:14.250 align:middle line:84% So that's the free rider problem in practice. 00:17:14.250 --> 00:17:19.130 align:middle line:90% The question is, can-- 00:17:19.130 --> 00:17:24.530 align:middle line:84% now the fact, the private market doesn't deliver optimality 00:17:24.530 --> 00:17:27.510 align:middle line:84% does not by itself say the government should be involved. 00:17:27.510 --> 00:17:30.870 align:middle line:84% In fact, once again, our friend Coase spoke up and said, 00:17:30.870 --> 00:17:33.440 align:middle line:84% wait a second, the private market 00:17:33.440 --> 00:17:34.900 align:middle line:90% can solve public good problems. 00:17:34.900 --> 00:17:39.000 align:middle line:84% And he pointed to a famous example of lighthouses. 00:17:39.000 --> 00:17:42.600 align:middle line:84% He said, a lighthouse seems to fit the classic definition 00:17:42.600 --> 00:17:44.200 align:middle line:90% of a pure public good. 00:17:44.200 --> 00:17:47.680 align:middle line:84% You can't exclude someone from seeing the beacon. 00:17:47.680 --> 00:17:50.000 align:middle line:84% And if the beacon is out there and I'm using, 00:17:50.000 --> 00:17:51.380 align:middle line:90% it doesn't deter your use of it. 00:17:51.380 --> 00:17:53.260 align:middle line:84% So lighthouses are a pretty pure public good. 00:17:53.260 --> 00:17:56.840 align:middle line:84% And yet for much of history, lighthouses 00:17:56.840 --> 00:17:58.680 align:middle line:90% were privately provided. 00:17:58.680 --> 00:18:00.820 align:middle line:84% There was no government provision of lighthouses. 00:18:00.820 --> 00:18:02.520 align:middle line:90% They were privately provided. 00:18:02.520 --> 00:18:03.340 align:middle line:90% How did they do it? 00:18:03.340 --> 00:18:04.840 align:middle line:84% When your boat landed, they wouldn't 00:18:04.840 --> 00:18:06.040 align:middle line:84% let you take the cargo off the boat 00:18:06.040 --> 00:18:08.440 align:middle line:84% until you paid a fee that offset the cost of running 00:18:08.440 --> 00:18:09.840 align:middle line:90% the lighthouse. 00:18:09.840 --> 00:18:13.160 align:middle line:84% So since they had a property right to docking, 00:18:13.160 --> 00:18:16.800 align:middle line:84% they could essentially, in a Coasean way, 00:18:16.800 --> 00:18:18.400 align:middle line:90% solve the public goods problem. 00:18:18.400 --> 00:18:20.735 align:middle line:84% So once again, in theory, the same theoretical Coasean 00:18:20.735 --> 00:18:22.360 align:middle line:84% argument, which says that externalities 00:18:22.360 --> 00:18:25.400 align:middle line:84% can be internalized, says public goods problems can be solved, 00:18:25.400 --> 00:18:27.760 align:middle line:84% as long as there's property rights. 00:18:27.760 --> 00:18:32.060 align:middle line:84% And indeed, a famous example of this occurring 00:18:32.060 --> 00:18:35.558 align:middle line:84% is something called business improvement districts. 00:18:35.558 --> 00:18:37.100 align:middle line:84% Once again, we don't have New Yorkers 00:18:37.100 --> 00:18:40.540 align:middle line:84% here, but how many of you guys have been to Times Square? 00:18:40.540 --> 00:18:42.620 align:middle line:90% OK, most of you. 00:18:42.620 --> 00:18:46.580 align:middle line:84% Times square, when I was a kid, was a nightmare. 00:18:46.580 --> 00:18:49.180 align:middle line:84% I mean, I can't believe my parents let me go there. 00:18:49.180 --> 00:18:53.240 align:middle line:84% It was just prostitutes and gambling and drugs. 00:18:53.240 --> 00:18:54.420 align:middle line:90% It was awful. 00:18:54.420 --> 00:18:57.820 align:middle line:84% Now it's just Disney characters and M&M stores. 00:18:57.820 --> 00:18:59.582 align:middle line:90% What happened? 00:18:59.582 --> 00:19:02.040 align:middle line:84% What happened was the government was failing to provide it. 00:19:02.040 --> 00:19:04.860 align:middle line:84% And New York had a law, which is called a Business Improvement 00:19:04.860 --> 00:19:05.860 align:middle line:90% District law. 00:19:05.860 --> 00:19:13.220 align:middle line:84% And the way this law worked was that basically, the law 00:19:13.220 --> 00:19:16.300 align:middle line:84% said that if you have an organization of businesses 00:19:16.300 --> 00:19:22.700 align:middle line:84% and 60% of them agree, you can levy a charge on every business 00:19:22.700 --> 00:19:24.787 align:middle line:90% for private purposes. 00:19:24.787 --> 00:19:27.120 align:middle line:84% So the businesses in Times Square got together and said, 00:19:27.120 --> 00:19:29.530 align:middle line:84% as long as 60% of us agree, we can 00:19:29.530 --> 00:19:31.490 align:middle line:84% force every business in Times Square 00:19:31.490 --> 00:19:34.050 align:middle line:84% to pay a charge that will go to private sanitation 00:19:34.050 --> 00:19:36.290 align:middle line:90% and private enforcement. 00:19:36.290 --> 00:19:38.710 align:middle line:84% And they did, and they privately cleaned up Times Square. 00:19:38.710 --> 00:19:40.710 align:middle line:84% The Times Square Business Improvement District's 00:19:40.710 --> 00:19:46.050 align:middle line:84% one of most successful examples in the country, where 00:19:46.050 --> 00:19:48.770 align:middle line:84% basically the budget of the Times Square Alliance 00:19:48.770 --> 00:19:51.590 align:middle line:84% is about $23 million, about 200 employees. 00:19:51.590 --> 00:19:54.290 align:middle line:84% And they keep the place clean and safe. 00:19:54.290 --> 00:19:56.830 align:middle line:90% So that's a Coasean solution. 00:19:56.830 --> 00:19:59.223 align:middle line:90% 00:19:59.223 --> 00:20:00.890 align:middle line:84% And these business improvement districts 00:20:00.890 --> 00:20:01.790 align:middle line:90% are all over the country. 00:20:01.790 --> 00:20:04.040 align:middle line:84% Almost every state has a law with business improvement 00:20:04.040 --> 00:20:04.850 align:middle line:90% districts. 00:20:04.850 --> 00:20:07.490 align:middle line:90% And they've grown. 00:20:07.490 --> 00:20:10.050 align:middle line:84% There's a big one in Los Angeles, which provided 00:20:10.050 --> 00:20:12.810 align:middle line:90% crime reductions et cetera 00:20:12.810 --> 00:20:15.970 align:middle line:84% Probably the state with the fewest BIDs, 00:20:15.970 --> 00:20:19.450 align:middle line:84% and the least successful, is Massachusetts. 00:20:19.450 --> 00:20:21.090 align:middle line:90% Now, why is that? 00:20:21.090 --> 00:20:24.250 align:middle line:84% Well, think about the key way the Massachusetts law differs 00:20:24.250 --> 00:20:33.040 align:middle line:84% from other laws, which is that if the vote passes in your area, 00:20:33.040 --> 00:20:37.240 align:middle line:84% within 30 days, you can opt out of paying the tax. 00:20:37.240 --> 00:20:41.620 align:middle line:84% So if the district in Waltham, which is a suburb of Boston, 00:20:41.620 --> 00:20:45.120 align:middle line:84% 60% of the businesses vote to tax themselves. 00:20:45.120 --> 00:20:48.640 align:middle line:84% The other 40% could just say, no thanks, we're not going to pay. 00:20:48.640 --> 00:20:49.960 align:middle line:90% Well, what does that do? 00:20:49.960 --> 00:20:52.120 align:middle line:84% Reintroduce the free rider problem, 00:20:52.120 --> 00:20:54.660 align:middle line:84% because they know once [INAUDIBLE] ride, 00:20:54.660 --> 00:20:55.680 align:middle line:90% they'll benefit. 00:20:55.680 --> 00:20:57.880 align:middle line:90% So why should they pay? 00:20:57.880 --> 00:21:01.293 align:middle line:84% So subsets, you need more than a property right 00:21:01.293 --> 00:21:02.710 align:middle line:84% to make the Coasean solution work. 00:21:02.710 --> 00:21:04.168 align:middle line:84% For business improvement districts, 00:21:04.168 --> 00:21:08.080 align:middle line:84% you need correct legislation to enforce the absence 00:21:08.080 --> 00:21:09.380 align:middle line:90% of the free rider problem. 00:21:09.380 --> 00:21:12.620 align:middle line:84% It was the fact that in Times Square, when 60% approved, 00:21:12.620 --> 00:21:15.080 align:middle line:84% everyone had to pay, make it work. 00:21:15.080 --> 00:21:21.000 align:middle line:84% In Massachusetts, when 60% approve, if people can drop out, 00:21:21.000 --> 00:21:23.080 align:middle line:84% then the whole thing falls apart. 00:21:23.080 --> 00:21:24.020 align:middle line:90% Yeah. 00:21:24.020 --> 00:21:27.860 align:middle line:84% AUDIENCE: Do the idea is that don't have that opt-out clause. 00:21:27.860 --> 00:21:31.060 align:middle line:84% It's like legal challenges where if someone owns a storefront 00:21:31.060 --> 00:21:34.120 align:middle line:84% that they operate [INAUDIBLE], who are the other business? 00:21:34.120 --> 00:21:36.100 align:middle line:90% Like, what legal authority-- 00:21:36.100 --> 00:21:37.880 align:middle line:84% JON GRUBER: It's a legal authority. 00:21:37.880 --> 00:21:39.500 align:middle line:90% It's a great question. 00:21:39.500 --> 00:21:41.000 align:middle line:84% I don't know enough of the legality. 00:21:41.000 --> 00:21:42.120 align:middle line:84% I don't know what lawsuits they face. 00:21:42.120 --> 00:21:44.460 align:middle line:84% They've certainly withstood any lawsuits they faced. 00:21:44.460 --> 00:21:47.380 align:middle line:84% But basically, the state has given a legal authority 00:21:47.380 --> 00:21:49.900 align:middle line:84% for businesses to essentially create one of these business 00:21:49.900 --> 00:21:51.080 align:middle line:90% improvement districts. 00:21:51.080 --> 00:21:53.310 align:middle line:84% And I presume the charges are nominal. 00:21:53.310 --> 00:21:55.060 align:middle line:84% If they were phenomenal, presumably people 00:21:55.060 --> 00:21:55.760 align:middle line:90% would sue a lot. 00:21:55.760 --> 00:21:57.302 align:middle line:84% But maybe they're nominal if it's not 00:21:57.302 --> 00:21:58.638 align:middle line:90% worth paying the lawyers to sue. 00:21:58.638 --> 00:21:59.680 align:middle line:90% But it's a good question. 00:21:59.680 --> 00:22:04.620 align:middle line:84% I don't know the legality of it, how it's been challenged. 00:22:04.620 --> 00:22:09.628 align:middle line:84% So in some sense, private provision has a case here. 00:22:09.628 --> 00:22:10.920 align:middle line:90% There's the lighthouse example. 00:22:10.920 --> 00:22:12.820 align:middle line:84% There's the business improvement districts. 00:22:12.820 --> 00:22:17.460 align:middle line:84% So the question is, when is private provision 00:22:17.460 --> 00:22:21.740 align:middle line:84% likely to actually work in providing public goods 00:22:21.740 --> 00:22:23.260 align:middle line:90% and when is it not? 00:22:23.260 --> 00:22:25.570 align:middle line:84% What are the conditions under which it will 00:22:25.570 --> 00:22:28.130 align:middle line:90% work, under which it will not? 00:22:28.130 --> 00:22:32.810 align:middle line:84% And in some sense, a key example, a key point, 00:22:32.810 --> 00:22:37.210 align:middle line:84% is private provision will work best 00:22:37.210 --> 00:22:42.370 align:middle line:84% when there's heterogeneity among the parties involved. 00:22:42.370 --> 00:22:47.650 align:middle line:84% The more heterogeneity in either income or tastes, 00:22:47.650 --> 00:22:52.090 align:middle line:84% the better private provision of public goods might work. 00:22:52.090 --> 00:22:54.570 align:middle line:90% Why is that? 00:22:54.570 --> 00:23:01.730 align:middle line:84% Imagine Ben and Jerry in their two-person economy, 00:23:01.730 --> 00:23:05.810 align:middle line:84% but Ben is very much richer than Jerry. 00:23:05.810 --> 00:23:09.210 align:middle line:84% In that case, we will get closer to the optimal provision 00:23:09.210 --> 00:23:12.330 align:middle line:84% of fireworks than if they have the same incomes. 00:23:12.330 --> 00:23:14.330 align:middle line:90% Why? 00:23:14.330 --> 00:23:16.890 align:middle line:84% And for an intuition, think about the extreme case. 00:23:16.890 --> 00:23:18.890 align:middle line:90% Imagine 00:23:18.890 --> 00:23:23.680 align:middle line:84% Ben has all the money, Jerry has no money. 00:23:23.680 --> 00:23:26.312 align:middle line:84% We still won't get to the optimal. 00:23:26.312 --> 00:23:27.020 align:middle line:90% We'll get closer. 00:23:27.020 --> 00:23:27.620 align:middle line:90% Why is that? 00:23:27.620 --> 00:23:31.880 align:middle line:90% 00:23:31.880 --> 00:23:33.900 align:middle line:84% Why does that heterogeneity-- or if you like, 00:23:33.900 --> 00:23:35.420 align:middle line:84% think about heterogeneity in taste. 00:23:35.420 --> 00:23:39.160 align:middle line:84% Imagine Ben likes fireworks a ton more than Jerry. 00:23:39.160 --> 00:23:40.940 align:middle line:84% You'll also get closer to the optimum. 00:23:40.940 --> 00:23:43.800 align:middle line:90% 00:23:43.800 --> 00:23:45.720 align:middle line:90% Well, think of it this way. 00:23:45.720 --> 00:23:49.720 align:middle line:84% Does Robinson Crusoe face a public goods problem when he's 00:23:49.720 --> 00:23:50.940 align:middle line:90% trapped alone in his island? 00:23:50.940 --> 00:23:52.360 align:middle line:90% You guys know Robinson Crusoe? 00:23:52.360 --> 00:23:53.540 align:middle line:84% You guys don't even know who he is. 00:23:53.540 --> 00:23:54.140 align:middle line:90% It's an old story. 00:23:54.140 --> 00:23:55.140 align:middle line:84% He got trapped alone in an island. 00:23:55.140 --> 00:23:56.320 align:middle line:90% That's all you need to know. 00:23:56.320 --> 00:23:58.020 align:middle line:84% He doesn't have a public good problem. 00:23:58.020 --> 00:23:59.320 align:middle line:90% Why? 00:23:59.320 --> 00:24:03.480 align:middle line:84% He has a free rider problem because it's just him. 00:24:03.480 --> 00:24:07.560 align:middle line:84% As most of the resources and most 00:24:07.560 --> 00:24:13.640 align:middle line:84% of the taste for public goods get coordinated into one person, 00:24:13.640 --> 00:24:16.360 align:middle line:84% it gets closer to Robinson Crusoe. 00:24:16.360 --> 00:24:18.727 align:middle line:84% So you get closer to the optimum. 00:24:18.727 --> 00:24:20.560 align:middle line:84% So let's see that mathematically, once again 00:24:20.560 --> 00:24:22.420 align:middle line:90% from the appendix. 00:24:22.420 --> 00:24:26.820 align:middle line:84% Imagine that Ben now has an income of $125, 00:24:26.820 --> 00:24:28.920 align:middle line:90% and Jerry's income is only $75. 00:24:28.920 --> 00:24:35.813 align:middle line:84% So now yb is 125 and yj equals 75. 00:24:35.813 --> 00:24:37.480 align:middle line:84% We've redistributed income a little bit, 00:24:37.480 --> 00:24:39.860 align:middle line:84% so now Ben has more money but the utilities 00:24:39.860 --> 00:24:41.860 align:middle line:90% are still the same. 00:24:41.860 --> 00:24:47.080 align:middle line:84% Now Ben has a demand for fireworks. 00:24:47.080 --> 00:24:49.500 align:middle line:84% So if you do the same math, f sub B, 00:24:49.500 --> 00:24:51.540 align:middle line:84% Ben's demand for fireworks is going 00:24:51.540 --> 00:24:58.340 align:middle line:90% to be 125 minus f sub J over 3. 00:24:58.340 --> 00:25:06.820 align:middle line:84% And Jerry's is going to be 75 minus f sub B over 3. 00:25:06.820 --> 00:25:08.360 align:middle line:90% Sorry-- minus 2 f sub j. 00:25:08.360 --> 00:25:10.620 align:middle line:90% There's a 2 here. 00:25:10.620 --> 00:25:11.820 align:middle line:90% Do I have the 2 there? 00:25:11.820 --> 00:25:13.192 align:middle line:90% Did I screw that up? 00:25:13.192 --> 00:25:14.400 align:middle line:90% I might have screwed that up. 00:25:14.400 --> 00:25:14.780 align:middle line:90% Hold on. 00:25:14.780 --> 00:25:16.072 align:middle line:90% Yeah, there should be 2's here. 00:25:16.072 --> 00:25:16.940 align:middle line:90% That's my bad. 00:25:16.940 --> 00:25:18.240 align:middle line:90% There should be 2's here. 00:25:18.240 --> 00:25:20.130 align:middle line:90% This makes the math work. 00:25:20.130 --> 00:25:20.828 align:middle line:90% OK, I'm sorry. 00:25:20.828 --> 00:25:21.870 align:middle line:90% It's all in the appendix. 00:25:21.870 --> 00:25:23.245 align:middle line:84% But if you go back to your notes, 00:25:23.245 --> 00:25:25.170 align:middle line:84% there should be 2's in front of those. 00:25:25.170 --> 00:25:26.798 align:middle line:90% Yeah, I got it right up there. 00:25:26.798 --> 00:25:28.090 align:middle line:90% But I didn't get it right here. 00:25:28.090 --> 00:25:30.076 align:middle line:90% This should be 2's here. 00:25:30.076 --> 00:25:34.370 align:middle line:84% Even when I'm reading out of the textbook, I can't get it right. 00:25:34.370 --> 00:25:36.890 align:middle line:90% You guys see that? 00:25:36.890 --> 00:25:45.050 align:middle line:84% OK, so basically, it's 125 minus 2 fj over 75 minus 2 fB over 3. 00:25:45.050 --> 00:25:47.730 align:middle line:84% When you solve those two equations jointly, 00:25:47.730 --> 00:25:53.290 align:middle line:84% you're going to get that f sub B equals 45 00:25:53.290 --> 00:25:58.730 align:middle line:90% and f sub j equals minus 5. 00:25:58.730 --> 00:26:00.250 align:middle line:84% That's what you're going to get when 00:26:00.250 --> 00:26:03.050 align:middle line:84% you solve those equations f sub B equals 45 00:26:03.050 --> 00:26:06.250 align:middle line:90% and f sub j equals minus 5. 00:26:06.250 --> 00:26:11.210 align:middle line:84% Now, you can't provide negative fireworks. 00:26:11.210 --> 00:26:15.570 align:middle line:90% So f sub j becomes 0. 00:26:15.570 --> 00:26:18.510 align:middle line:84% So the total amount of fireworks provided becomes 45. 00:26:18.510 --> 00:26:24.400 align:middle line:90% 00:26:24.400 --> 00:26:29.240 align:middle line:84% Yeah, and the total amount of fireworks becomes 45. 00:26:29.240 --> 00:26:34.040 align:middle line:84% This is higher than what they get when-- 00:26:34.040 --> 00:26:37.200 align:middle line:84% no, actually, it doesn't become 45, because once this is 0, 00:26:37.200 --> 00:26:38.720 align:middle line:90% you re-solve for Ben. 00:26:38.720 --> 00:26:42.260 align:middle line:84% And once that's re-solved for Ben, 00:26:42.260 --> 00:26:47.602 align:middle line:84% you actually get that f, which is f sub B equals 41.66. 00:26:47.602 --> 00:26:49.560 align:middle line:84% Because once this is 0, you got to re-plug this 00:26:49.560 --> 00:26:51.200 align:middle line:90% into Ben's condition. 00:26:51.200 --> 00:26:53.940 align:middle line:84% And you're going to get that the total amount of fireworks fired 00:26:53.940 --> 00:26:55.502 align:middle line:90% is 41.66. 00:26:55.502 --> 00:26:57.960 align:middle line:84% So if you solve it, you get a solution that can't be right, 00:26:57.960 --> 00:27:00.020 align:middle line:84% which is Jerry can't provide negative fireworks. 00:27:00.020 --> 00:27:01.060 align:middle line:90% Jerry provides 0. 00:27:01.060 --> 00:27:03.940 align:middle line:84% You plug 0 back in here, and you get that Ben's fireworks-- 00:27:03.940 --> 00:27:07.320 align:middle line:84% Ben's the only person provide fireworks, is 41.66. 00:27:07.320 --> 00:27:09.080 align:middle line:90% Take this to the extreme. 00:27:09.080 --> 00:27:11.563 align:middle line:84% If Ben has all the money and Jerry has none of the money, 00:27:11.563 --> 00:27:13.980 align:middle line:84% you're going to get very, very close to the optimal level. 00:27:13.980 --> 00:27:16.480 align:middle line:90% You'll never get there. 00:27:16.480 --> 00:27:18.300 align:middle line:84% The way I like to think is the following. 00:27:18.300 --> 00:27:22.140 align:middle line:84% Imagine you have a giant mansion and a tiny crappy 00:27:22.140 --> 00:27:26.960 align:middle line:90% shack that share a driveway. 00:27:26.960 --> 00:27:31.033 align:middle line:84% Maybe there's a little extra bit that goes off to each. 00:27:31.033 --> 00:27:32.700 align:middle line:84% There should technically be a free rider 00:27:32.700 --> 00:27:35.380 align:middle line:84% problem in plowing that driveway, 00:27:35.380 --> 00:27:38.300 align:middle line:84% but the mansion is just going to fricking plow the driveway. 00:27:38.300 --> 00:27:43.040 align:middle line:84% They're not going to not plow it because the shack won't kick in. 00:27:43.040 --> 00:27:44.918 align:middle line:84% Now, maybe they won't do quite as good a job 00:27:44.918 --> 00:27:46.460 align:middle line:84% as they'd do if the shack was paying. 00:27:46.460 --> 00:27:48.043 align:middle line:84% They certainly won't do the little bit 00:27:48.043 --> 00:27:49.340 align:middle line:90% that goes off to the shack. 00:27:49.340 --> 00:27:51.860 align:middle line:84% But they're going to pretty much plow the driveway. 00:27:51.860 --> 00:27:54.700 align:middle line:84% The point is, the more one party really 00:27:54.700 --> 00:27:56.537 align:middle line:84% has all the stake in the decision, 00:27:56.537 --> 00:27:58.620 align:middle line:84% the more you move away from the free rider problem 00:27:58.620 --> 00:28:01.340 align:middle line:84% towards essentially the Robinson Crusoe solution, which is 00:28:01.340 --> 00:28:02.560 align:middle line:90% they just do the right thing. 00:28:02.560 --> 00:28:03.060 align:middle line:90% Yeah. 00:28:03.060 --> 00:28:04.780 align:middle line:84% AUDIENCE: When introducing heterogeneity, 00:28:04.780 --> 00:28:08.220 align:middle line:84% does the distribution of resources and tastes 00:28:08.220 --> 00:28:10.380 align:middle line:84% have to be positively correlated? 00:28:10.380 --> 00:28:14.580 align:middle line:84% So say, for example, one party has all the resources but none 00:28:14.580 --> 00:28:15.200 align:middle line:90% of the taste. 00:28:15.200 --> 00:28:16.070 align:middle line:90% JON GRUBER: Yes, that's right. 00:28:16.070 --> 00:28:16.903 align:middle line:90% That's a good point. 00:28:16.903 --> 00:28:19.250 align:middle line:84% That could be offsetting heterogeneity. 00:28:19.250 --> 00:28:20.650 align:middle line:90% That's a great point. 00:28:20.650 --> 00:28:24.533 align:middle line:84% I've assumed that there's no correlation here. 00:28:24.533 --> 00:28:25.950 align:middle line:84% They both like fireworks the same. 00:28:25.950 --> 00:28:28.367 align:middle line:84% But if there's offsetting homogeneity, that would undo it. 00:28:28.367 --> 00:28:29.490 align:middle line:90% Great point. 00:28:29.490 --> 00:28:32.610 align:middle line:90% Other comments or questions? 00:28:32.610 --> 00:28:35.170 align:middle line:90% So that's one-- yeah. 00:28:35.170 --> 00:28:39.610 align:middle line:84% AUDIENCE: Among [INAUDIBLE] since we're 00:28:39.610 --> 00:28:41.890 align:middle line:84% talking about heterogeneity, is there 00:28:41.890 --> 00:28:43.470 align:middle line:84% a hard line between heterogeneity 00:28:43.470 --> 00:28:46.182 align:middle line:84% and just overall concentration of resources? 00:28:46.182 --> 00:28:47.890 align:middle line:84% JON GRUBER: Yeah, concentration is really 00:28:47.890 --> 00:28:50.850 align:middle line:90% going to make the difference. 00:28:50.850 --> 00:28:52.490 align:middle line:90% So that's one reason. 00:28:52.490 --> 00:28:55.330 align:middle line:84% So we're talking about, when is the private sector 00:28:55.330 --> 00:28:57.810 align:middle line:84% going to be closer to getting it right? 00:28:57.810 --> 00:29:00.207 align:middle line:84% One is when there's this concentration. 00:29:00.207 --> 00:29:02.290 align:middle line:84% That's a better term than heterogeneity, actually. 00:29:02.290 --> 00:29:04.570 align:middle line:90% When this is "concentration." 00:29:04.570 --> 00:29:08.210 align:middle line:84% The second is when there's altruism. 00:29:08.210 --> 00:29:16.400 align:middle line:84% Altruism is defined as the fact that my utility includes 00:29:16.400 --> 00:29:18.080 align:middle line:90% your utility. 00:29:18.080 --> 00:29:19.680 align:middle line:84% An altruistic utility function is 00:29:19.680 --> 00:29:25.400 align:middle line:84% a function of my own consumption plus your utility. 00:29:25.400 --> 00:29:28.320 align:middle line:84% That's an altruistic utility function. 00:29:28.320 --> 00:29:31.040 align:middle line:84% If individuals are altruistic, then that 00:29:31.040 --> 00:29:34.360 align:middle line:84% helps solve the free rider problem, 00:29:34.360 --> 00:29:36.240 align:middle line:84% because they're incorporating the fact 00:29:36.240 --> 00:29:39.400 align:middle line:84% that the other person would be happier with more fireworks. 00:29:39.400 --> 00:29:41.520 align:middle line:84% Once again, it doesn't fully solve it 00:29:41.520 --> 00:29:43.180 align:middle line:90% except under certain conditions. 00:29:43.180 --> 00:29:46.560 align:middle line:84% But it will move us closer to the optimum. 00:29:46.560 --> 00:29:51.660 align:middle line:84% Now, it turns out that, much to the chagrin of economists, 00:29:51.660 --> 00:29:53.780 align:middle line:90% people are pretty altruistic. 00:29:53.780 --> 00:29:55.823 align:middle line:84% It's not in our standard models we teach. 00:29:55.823 --> 00:29:57.740 align:middle line:84% But it turns out people are pretty altruistic. 00:29:57.740 --> 00:30:01.120 align:middle line:84% And the best way to see this is to look at the dictator game. 00:30:01.120 --> 00:30:04.292 align:middle line:84% Any of you guys ever been to the science museum in Boston? 00:30:04.292 --> 00:30:06.000 align:middle line:84% You probably didn't go there, but there's 00:30:06.000 --> 00:30:08.812 align:middle line:84% a room where there's computers that can play a game where you 00:30:08.812 --> 00:30:10.020 align:middle line:90% trade cookies back and forth. 00:30:10.020 --> 00:30:11.273 align:middle line:90% Any you guys ever play that? 00:30:11.273 --> 00:30:12.440 align:middle line:90% I used to play with my kids. 00:30:12.440 --> 00:30:15.080 align:middle line:84% OK, it's more of a kids thing anyway. 00:30:15.080 --> 00:30:20.260 align:middle line:84% So basically, imagine the following experiment. 00:30:20.260 --> 00:30:23.030 align:middle line:84% It's a typical experiment, run hundreds of times 00:30:23.030 --> 00:30:25.820 align:middle line:84% at universities all over the country. 00:30:25.820 --> 00:30:28.060 align:middle line:84% You have five students who are going 00:30:28.060 --> 00:30:30.060 align:middle line:90% to play 10 rounds of a game. 00:30:30.060 --> 00:30:32.760 align:middle line:84% And the way the game is going to work is, in each round, 00:30:32.760 --> 00:30:35.220 align:middle line:90% students are given $1. 00:30:35.220 --> 00:30:37.900 align:middle line:84% And they have the option of keeping that dollar 00:30:37.900 --> 00:30:40.400 align:middle line:84% or putting it in the public fund, which is in the middle. 00:30:40.400 --> 00:30:43.010 align:middle line:90% 00:30:43.010 --> 00:30:44.620 align:middle line:90% If they could choose-- 00:30:44.620 --> 00:30:48.820 align:middle line:90% 00:30:48.820 --> 00:30:51.580 align:middle line:84% after all students decide whether to contribute, 00:30:51.580 --> 00:30:54.060 align:middle line:84% the amount in the public fund is then 00:30:54.060 --> 00:30:57.340 align:middle line:84% split up equally and distributed back to everyone. 00:30:57.340 --> 00:30:58.700 align:middle line:90% So everyone gets $1. 00:30:58.700 --> 00:31:00.860 align:middle line:90% These five guys get $1. 00:31:00.860 --> 00:31:02.460 align:middle line:84% They have a choice, keep that dollar 00:31:02.460 --> 00:31:04.040 align:middle line:90% or put it in a public fund. 00:31:04.040 --> 00:31:05.620 align:middle line:84% If they put it in the public fund, 00:31:05.620 --> 00:31:10.090 align:middle line:84% it gets doubled and distributed equally back to everyone. 00:31:10.090 --> 00:31:17.710 align:middle line:84% So in a perfectly coordinated cooperative equilibrium, 00:31:17.710 --> 00:31:20.210 align:middle line:84% what's the optimal thing to do for these guys? 00:31:20.210 --> 00:31:20.870 align:middle line:90% Yeah. 00:31:20.870 --> 00:31:21.810 align:middle line:84% AUDIENCE: Everyone puts in their dollar. 00:31:21.810 --> 00:31:24.352 align:middle line:84% JON GRUBER: Puts in their dollar and they all end up with $2. 00:31:24.352 --> 00:31:30.008 align:middle line:84% In a Nash equilibrium, what should each person do? 00:31:30.008 --> 00:31:30.550 align:middle line:90% Someone else. 00:31:30.550 --> 00:31:33.170 align:middle line:90% What should each person do? 00:31:33.170 --> 00:31:34.910 align:middle line:90% What's privately optimal? 00:31:34.910 --> 00:31:36.690 align:middle line:84% Socially optimal if you each put in $1. 00:31:36.690 --> 00:31:38.705 align:middle line:90% What's privately optimal? 00:31:38.705 --> 00:31:39.830 align:middle line:90% Let's get some other folks. 00:31:39.830 --> 00:31:40.090 align:middle line:90% Yeah. 00:31:40.090 --> 00:31:41.170 align:middle line:90% AUDIENCE: Keeping their money. 00:31:41.170 --> 00:31:43.712 align:middle line:84% JON GRUBER: Keeping their money, because every dollar you put 00:31:43.712 --> 00:31:46.770 align:middle line:90% in, you get $0.40 back. 00:31:46.770 --> 00:31:49.630 align:middle line:84% So why would you put in your dollar? 00:31:49.630 --> 00:31:52.030 align:middle line:90% So there's a free rider problem. 00:31:52.030 --> 00:31:55.610 align:middle line:84% In practice, you put in $1, but you only get $0.40 back of that 00:31:55.610 --> 00:31:57.170 align:middle line:90% dollar. 00:31:57.170 --> 00:31:58.770 align:middle line:90% So why would you put it in? 00:31:58.770 --> 00:32:00.770 align:middle line:84% Well, in fact, when you run these games, when 00:32:00.770 --> 00:32:03.530 align:middle line:84% you run these experiments in practice, 00:32:03.530 --> 00:32:11.120 align:middle line:84% you find that roughly half of participants 00:32:11.120 --> 00:32:12.680 align:middle line:90% put in their dollar. 00:32:12.680 --> 00:32:16.600 align:middle line:84% Even though there is no private incentive, roughly half 00:32:16.600 --> 00:32:19.480 align:middle line:84% put in the dollar, which is consistent 00:32:19.480 --> 00:32:22.840 align:middle line:84% with altruism, consistent with the notion that basically-- now, 00:32:22.840 --> 00:32:25.060 align:middle line:84% altruism could arrive through social appropriation, 00:32:25.060 --> 00:32:27.600 align:middle line:84% like I feel like a jerk if I don't put in my dollar. 00:32:27.600 --> 00:32:31.040 align:middle line:84% It could arise through true loving your fellow person. 00:32:31.040 --> 00:32:33.640 align:middle line:84% However it arises, it effectively 00:32:33.640 --> 00:32:36.740 align:middle line:84% acts as if you care about your fellow person. 00:32:36.740 --> 00:32:38.843 align:middle line:84% It operationalizes itself in the same way. 00:32:38.843 --> 00:32:41.260 align:middle line:84% It's an interesting question about where it comes through, 00:32:41.260 --> 00:32:43.760 align:middle line:84% but that's irrelevant to the reduced form outcome, which 00:32:43.760 --> 00:32:49.320 align:middle line:84% sort of operationalizes itself as people being altruistic. 00:32:49.320 --> 00:32:49.820 align:middle line:90% Yeah. 00:32:49.820 --> 00:32:51.960 align:middle line:84% AUDIENCE: This might be slightly off topic, 00:32:51.960 --> 00:32:55.080 align:middle line:84% but would this be different than a lottery example 00:32:55.080 --> 00:32:57.920 align:middle line:84% because there's uncertainty in how many others 00:32:57.920 --> 00:33:01.493 align:middle line:90% contribute $1 when you-- 00:33:01.493 --> 00:33:03.385 align:middle line:84% if I should contribute $1 or not. 00:33:03.385 --> 00:33:05.260 align:middle line:84% JON GRUBER: Different from a lottery example? 00:33:05.260 --> 00:33:06.180 align:middle line:90% What do you mean? 00:33:06.180 --> 00:33:07.900 align:middle line:90% AUDIENCE: [INAUDIBLE]. 00:33:07.900 --> 00:33:10.240 align:middle line:84% So for example, the most optimal thing for me, 00:33:10.240 --> 00:33:12.340 align:middle line:84% if I could coordinate everyone else's behavior, 00:33:12.340 --> 00:33:16.240 align:middle line:84% is for me to keep my $1, but everyone else to contribute $1. 00:33:16.240 --> 00:33:18.460 align:middle line:84% JON GRUBER: No, the most optimal thing 00:33:18.460 --> 00:33:21.097 align:middle line:90% is you all put in your dollar. 00:33:21.097 --> 00:33:22.180 align:middle line:90% A lottery is not the same. 00:33:22.180 --> 00:33:23.980 align:middle line:90% There's only one winner. 00:33:23.980 --> 00:33:27.380 align:middle line:84% And as we learned in 14.01 actually, 00:33:27.380 --> 00:33:28.560 align:middle line:90% the lottery is the reverse. 00:33:28.560 --> 00:33:31.358 align:middle line:84% For every dollar you put in, you keep about $0.50. 00:33:31.358 --> 00:33:32.900 align:middle line:84% Because there's this multiplier, it's 00:33:32.900 --> 00:33:33.620 align:middle line:84% optimal for them all-- socially optimal 00:33:33.620 --> 00:33:34.953 align:middle line:90% for them all put their money in. 00:33:34.953 --> 00:33:37.220 align:middle line:84% And they're just making me pay more money. 00:33:37.220 --> 00:33:39.420 align:middle line:90% Privately, they don't do that. 00:33:39.420 --> 00:33:43.128 align:middle line:84% Now, it's quite interesting that there's that much-- 00:33:43.128 --> 00:33:44.920 align:middle line:84% there's some interesting lessons from that. 00:33:44.920 --> 00:33:47.920 align:middle line:84% First of all, you might say, well, gee, 00:33:47.920 --> 00:33:50.460 align:middle line:84% maybe people didn't understand it. 00:33:50.460 --> 00:33:52.460 align:middle line:84% How would you test whether it's about learning 00:33:52.460 --> 00:33:54.660 align:middle line:90% versus about altruism? 00:33:54.660 --> 00:33:57.420 align:middle line:84% How would you test that empirically? 00:33:57.420 --> 00:33:58.905 align:middle line:90% Someone tell me what you think. 00:33:58.905 --> 00:34:00.280 align:middle line:84% You're running these experiments. 00:34:00.280 --> 00:34:02.840 align:middle line:84% I say, yeah, you ran it, but maybe this don't understand it. 00:34:02.840 --> 00:34:03.938 align:middle line:90% What would you do? 00:34:03.938 --> 00:34:05.480 align:middle line:84% Someone raise their hand and tell me. 00:34:05.480 --> 00:34:08.090 align:middle line:90% 00:34:08.090 --> 00:34:08.670 align:middle line:90% Yeah. 00:34:08.670 --> 00:34:11.212 align:middle line:84% AUDIENCE: [INAUDIBLE] you could stop and explain [INAUDIBLE]. 00:34:11.212 --> 00:34:13.003 align:middle line:84% JON GRUBER: Or you could just play it again 00:34:13.003 --> 00:34:13.889 align:middle line:90% and again and again. 00:34:13.889 --> 00:34:16.592 align:middle line:84% That's the way we think economic intuition's developed. 00:34:16.592 --> 00:34:18.550 align:middle line:84% It turns out, if you played it again and again, 00:34:18.550 --> 00:34:20.429 align:middle line:84% people become a little less altruistic. 00:34:20.429 --> 00:34:23.330 align:middle line:84% But even after 10 rounds, people are very altruistic. 00:34:23.330 --> 00:34:26.190 align:middle line:90% So it's not a learning thing. 00:34:26.190 --> 00:34:28.690 align:middle line:84% Now, I don't know if they've tried explaining like, hey, you 00:34:28.690 --> 00:34:29.830 align:middle line:90% should be selfish. 00:34:29.830 --> 00:34:31.788 align:middle line:84% But they've had people play it again and again, 00:34:31.788 --> 00:34:33.429 align:middle line:84% and it doesn't make altruism go away. 00:34:33.429 --> 00:34:37.130 align:middle line:90% 00:34:37.130 --> 00:34:39.570 align:middle line:84% Now, of course, that's just a laboratory experiment. 00:34:39.570 --> 00:34:42.873 align:middle line:84% The question is, does altruism exist in the real world? 00:34:42.873 --> 00:34:44.290 align:middle line:84% And there's one great paper-- it's 00:34:44.290 --> 00:34:46.650 align:middle line:84% an older paper, but a great paper, which 00:34:46.650 --> 00:34:50.010 align:middle line:84% looked at contributions to public radio. 00:34:50.010 --> 00:34:52.510 align:middle line:84% It basically looked at people's giving, which is once again, 00:34:52.510 --> 00:34:55.110 align:middle line:84% a classic example of public good free rider problem-- 00:34:55.110 --> 00:34:56.469 align:middle line:84% you don't have to contribute to public radio. 00:34:56.469 --> 00:34:57.150 align:middle line:90% They make you feel guilty. 00:34:57.150 --> 00:34:58.870 align:middle line:84% But they don't come to your house or anything, 00:34:58.870 --> 00:35:00.210 align:middle line:90% or they don't twist your arm. 00:35:00.210 --> 00:35:03.050 align:middle line:84% You don't have to give to public radio. 00:35:03.050 --> 00:35:06.240 align:middle line:84% And what a typical economic model would say 00:35:06.240 --> 00:35:10.760 align:middle line:84% is, if I enjoy public radio, then as the number of listeners 00:35:10.760 --> 00:35:14.440 align:middle line:84% grows, I should be less likely to give, 00:35:14.440 --> 00:35:16.260 align:middle line:84% because basically, it's a fixed amount. 00:35:16.260 --> 00:35:18.400 align:middle line:84% Presuming the cost of running public radio 00:35:18.400 --> 00:35:20.740 align:middle line:84% is roughly the same as the number of listeners grows. 00:35:20.740 --> 00:35:22.720 align:middle line:84% It's just, you're sending the transmission out. 00:35:22.720 --> 00:35:24.178 align:middle line:84% In other words, as this grows, then 00:35:24.178 --> 00:35:26.640 align:middle line:84% basically there's more and more of a free rider problem. 00:35:26.640 --> 00:35:28.120 align:middle line:90% So think of it this way. 00:35:28.120 --> 00:35:30.840 align:middle line:84% Think of yourself, when you're coming out of the subway 00:35:30.840 --> 00:35:33.198 align:middle line:84% and there's no one else around, there's the guy begging, 00:35:33.198 --> 00:35:35.240 align:middle line:84% you feel more likely to get money than if there's 00:35:35.240 --> 00:35:36.800 align:middle line:90% 20 people walking by. 00:35:36.800 --> 00:35:37.983 align:middle line:90% It's just human nature. 00:35:37.983 --> 00:35:39.400 align:middle line:84% Likewise, it would say that people 00:35:39.400 --> 00:35:42.220 align:middle line:84% should give less to public radio as the listener base goes up. 00:35:42.220 --> 00:35:46.240 align:middle line:84% And in fact, they don't, which is consistent with the fact 00:35:46.240 --> 00:35:49.095 align:middle line:84% that there's some altruistic preferences going on. 00:35:49.095 --> 00:35:51.220 align:middle line:84% So that's one example of a real world test of this. 00:35:51.220 --> 00:35:52.840 align:middle line:90% There's lots more. 00:35:52.840 --> 00:35:59.000 align:middle line:84% Now, the problem is, then you ask the question of, well, 00:35:59.000 --> 00:36:00.880 align:middle line:90% where does altruism come from? 00:36:00.880 --> 00:36:02.060 align:middle line:90% Why are people altruistic? 00:36:02.060 --> 00:36:03.900 align:middle line:84% Is this just inherent in our nature? 00:36:03.900 --> 00:36:06.560 align:middle line:84% And it turns out altruism is not purely inherent, 00:36:06.560 --> 00:36:08.060 align:middle line:90% but can be influenced. 00:36:08.060 --> 00:36:09.860 align:middle line:84% And the main thing that influences altruism 00:36:09.860 --> 00:36:11.560 align:middle line:90% is the degree of social capital. 00:36:11.560 --> 00:36:15.900 align:middle line:90% 00:36:15.900 --> 00:36:17.260 align:middle line:90% This is the notion. 00:36:17.260 --> 00:36:19.360 align:middle line:84% Social capital is an abstract concept. 00:36:19.360 --> 00:36:22.060 align:middle line:84% But it's roughly the idea of how much you feel connected 00:36:22.060 --> 00:36:25.095 align:middle line:84% to your fellow people, how much you trust them, 00:36:25.095 --> 00:36:26.720 align:middle line:84% how much you feel integrated with them. 00:36:26.720 --> 00:36:28.387 align:middle line:84% And it turns out, if you run experiments 00:36:28.387 --> 00:36:31.460 align:middle line:84% in more connected societies, they're more altruistic. 00:36:31.460 --> 00:36:34.140 align:middle line:84% When you're playing with friends, you're more altruistic. 00:36:34.140 --> 00:36:35.835 align:middle line:84% So clearly, part of what drives altruism 00:36:35.835 --> 00:36:36.960 align:middle line:90% is actually knowing people. 00:36:36.960 --> 00:36:38.320 align:middle line:84% But even when you don't know people, 00:36:38.320 --> 00:36:39.720 align:middle line:90% there's still a lot of altruism. 00:36:39.720 --> 00:36:42.500 align:middle line:90% 00:36:42.500 --> 00:36:44.860 align:middle line:90% So that is another reason. 00:36:44.860 --> 00:36:48.060 align:middle line:84% We're talking about reasons why the private market might get 00:36:48.060 --> 00:36:49.860 align:middle line:90% closer to the public optimum. 00:36:49.860 --> 00:36:52.740 align:middle line:84% The private market can never, except in some extremely limited 00:36:52.740 --> 00:36:54.880 align:middle line:84% case, achieve the social optimum. 00:36:54.880 --> 00:36:58.260 align:middle line:84% But it might get closer if there's concentration. 00:36:58.260 --> 00:37:02.650 align:middle line:84% It might get closer if there is altruism. 00:37:02.650 --> 00:37:04.610 align:middle line:84% And the third thing that might drive it closer 00:37:04.610 --> 00:37:08.770 align:middle line:84% is what we call the warm glow model of giving. 00:37:08.770 --> 00:37:11.310 align:middle line:90% Now, this is like altruism. 00:37:11.310 --> 00:37:13.690 align:middle line:84% It has the same net result, but an important modeling 00:37:13.690 --> 00:37:15.050 align:middle line:90% difference. 00:37:15.050 --> 00:37:19.570 align:middle line:84% The warm glow model is, not that I care about you, 00:37:19.570 --> 00:37:22.875 align:middle line:84% but that I get utility from contributing. 00:37:22.875 --> 00:37:24.250 align:middle line:84% So in the fireworks, literally, I 00:37:24.250 --> 00:37:26.170 align:middle line:84% get jollies from lighting them off. 00:37:26.170 --> 00:37:27.590 align:middle line:90% It's literally a warm glow. 00:37:27.590 --> 00:37:29.153 align:middle line:84% It's not where it then comes from. 00:37:29.153 --> 00:37:31.570 align:middle line:84% But literally, if I get some utility was setting them off, 00:37:31.570 --> 00:37:36.410 align:middle line:84% then in my utility function is your fireworks 00:37:36.410 --> 00:37:37.410 align:middle line:90% because of that utility. 00:37:37.410 --> 00:37:41.410 align:middle line:84% Or more generally, if I get utility-- 00:37:41.410 --> 00:37:46.930 align:middle line:84% this is why churches and temples are filled with plaques 00:37:46.930 --> 00:37:49.150 align:middle line:84% and everybody's name is everywhere. 00:37:49.150 --> 00:37:50.650 align:middle line:84% People give partly because they want 00:37:50.650 --> 00:37:53.890 align:middle line:90% to see their name on a plaque. 00:37:53.890 --> 00:37:57.390 align:middle line:84% If I give, because I want to see my name on a plaque, 00:37:57.390 --> 00:37:59.050 align:middle line:90% that is not altruism. 00:37:59.050 --> 00:38:01.880 align:middle line:84% I literally get utility of giving. 00:38:01.880 --> 00:38:03.960 align:middle line:90% And that's the warm glow model. 00:38:03.960 --> 00:38:05.900 align:middle line:84% Now, it ends up with similar outcomes, 00:38:05.900 --> 00:38:08.880 align:middle line:90% but it's a very different model. 00:38:08.880 --> 00:38:12.160 align:middle line:84% The altruistic utility function is 00:38:12.160 --> 00:38:15.960 align:middle line:84% one where your utilities and my utility function. 00:38:15.960 --> 00:38:18.760 align:middle line:84% The warm glow utility function, is where my giving 00:38:18.760 --> 00:38:21.320 align:middle line:90% is in my utility function. 00:38:21.320 --> 00:38:25.480 align:middle line:84% Very different-- I get utility from my giving. 00:38:25.480 --> 00:38:30.240 align:middle line:84% It's similar effects but different models. 00:38:30.240 --> 00:38:37.620 align:middle line:84% So that's basically the bottom line on private provision. 00:38:37.620 --> 00:38:40.480 align:middle line:84% Private provision will never achieve the first best. 00:38:40.480 --> 00:38:45.280 align:middle line:84% But in certain conditions, it can get close Questions 00:38:45.280 --> 00:38:47.400 align:middle line:90% about that? 00:38:47.400 --> 00:38:48.240 align:middle line:90% Yeah, Enoch. 00:38:48.240 --> 00:38:50.000 align:middle line:84% AUDIENCE: How similar in this context 00:38:50.000 --> 00:38:52.520 align:middle line:84% is altruism from having an information, 00:38:52.520 --> 00:38:54.640 align:middle line:84% I guess, asymmetry, in the sense that if I 00:38:54.640 --> 00:38:58.370 align:middle line:84% know what my neighbor's utility function is, I guess, right? 00:38:58.370 --> 00:39:00.120 align:middle line:84% JON GRUBER: No, but if I'm not altruistic, 00:39:00.120 --> 00:39:01.547 align:middle line:90% I would exploit that. 00:39:01.547 --> 00:39:02.880 align:middle line:90% That wouldn't solve the problem. 00:39:02.880 --> 00:39:05.580 align:middle line:84% That would make it worse at one level. 00:39:05.580 --> 00:39:08.417 align:middle line:90% But no, that's actually is here. 00:39:08.417 --> 00:39:10.500 align:middle line:84% So here, I do know my neighbor's utility function. 00:39:10.500 --> 00:39:12.620 align:middle line:90% So imagine I didn't know. 00:39:12.620 --> 00:39:15.160 align:middle line:84% Well, that would just introduce uncertainty to this model. 00:39:15.160 --> 00:39:16.820 align:middle line:84% I'd be guessing, then, in which way-- 00:39:16.820 --> 00:39:17.820 align:middle line:84% I don't know which way I'm biased. 00:39:17.820 --> 00:39:18.695 align:middle line:90% That would affect it. 00:39:18.695 --> 00:39:21.300 align:middle line:84% But it wouldn't change the fundamental underprovision. 00:39:21.300 --> 00:39:22.300 align:middle line:90% Good question. 00:39:22.300 --> 00:39:23.560 align:middle line:90% Other questions about this? 00:39:23.560 --> 00:39:26.220 align:middle line:90% 00:39:26.220 --> 00:39:27.940 align:middle line:84% OK, so if the private market is not 00:39:27.940 --> 00:39:30.700 align:middle line:84% going to get us to the optimum, then we 00:39:30.700 --> 00:39:34.380 align:middle line:84% go to the next question on public finance, which is 00:39:34.380 --> 00:39:37.140 align:middle line:90% what should the government do? 00:39:37.140 --> 00:39:41.100 align:middle line:84% Or how should the government intervene? 00:39:41.100 --> 00:39:44.980 align:middle line:84% Well, before I get to that, what we're going to do-- here's 00:39:44.980 --> 00:39:47.860 align:middle line:84% the way that this sequence of lectures is going to work, 00:39:47.860 --> 00:39:51.793 align:middle line:84% which this is an overview of public goods. 00:39:51.793 --> 00:39:53.460 align:middle line:84% And then the next three lectures will be 00:39:53.460 --> 00:39:56.290 align:middle line:90% applications of public goods. 00:39:56.290 --> 00:39:59.650 align:middle line:84% So this overview, I'm going to talk 00:39:59.650 --> 00:40:03.490 align:middle line:84% about the three fundamental challenges facing 00:40:03.490 --> 00:40:06.370 align:middle line:84% public provision of public goods. 00:40:06.370 --> 00:40:09.743 align:middle line:84% In the next three lectures we'll talk about those challenges. 00:40:09.743 --> 00:40:11.410 align:middle line:84% So in other words, what I'm going to say 00:40:11.410 --> 00:40:14.230 align:middle line:84% is-- this is the key lesson from this class, 00:40:14.230 --> 00:40:16.370 align:middle line:90% lest we be too liberal. 00:40:16.370 --> 00:40:19.970 align:middle line:84% Market failure means that government has the potential 00:40:19.970 --> 00:40:24.050 align:middle line:84% to fix things, but not that government necessarily will. 00:40:24.050 --> 00:40:26.033 align:middle line:84% And what I talk about here, so far, government 00:40:26.033 --> 00:40:26.950 align:middle line:90% has been the good guy. 00:40:26.950 --> 00:40:28.550 align:middle line:84% We have an externality, someone comes in and fixes it. 00:40:28.550 --> 00:40:29.450 align:middle line:90% Yay. 00:40:29.450 --> 00:40:31.748 align:middle line:84% But in fact, that's not always true. 00:40:31.748 --> 00:40:33.290 align:middle line:84% And so we're going to talk about here 00:40:33.290 --> 00:40:37.330 align:middle line:84% is reasons why public provision may 00:40:37.330 --> 00:40:40.850 align:middle line:90% fail to get us to the optimum. 00:40:40.850 --> 00:40:46.270 align:middle line:84% And there's three reasons why public provision may fail. 00:40:46.270 --> 00:40:48.870 align:middle line:84% The first is the problem of crowd out. 00:40:48.870 --> 00:40:51.970 align:middle line:90% 00:40:51.970 --> 00:40:56.510 align:middle line:84% The first is the problem of crowd out, which 00:40:56.510 --> 00:41:01.070 align:middle line:84% is when the government comes in, part of what it will do 00:41:01.070 --> 00:41:03.750 align:middle line:84% is just displace what people are already 00:41:03.750 --> 00:41:06.830 align:middle line:84% doing, rather than actually increasing 00:41:06.830 --> 00:41:09.790 align:middle line:90% the amount of the public good. 00:41:09.790 --> 00:41:12.750 align:middle line:84% So to see that, let's make three assumptions 00:41:12.750 --> 00:41:15.110 align:middle line:90% in the firework example. 00:41:15.110 --> 00:41:17.550 align:middle line:84% First of all, let's assume there's 00:41:17.550 --> 00:41:19.350 align:middle line:90% no warm glow or altruism. 00:41:19.350 --> 00:41:21.510 align:middle line:84% Ben and Jerry just care about total f. 00:41:21.510 --> 00:41:25.230 align:middle line:84% They don't care who contributes, no warm glow of altruism. 00:41:25.230 --> 00:41:29.350 align:middle line:84% Second of all, let's say that in this symmetric example, 00:41:29.350 --> 00:41:32.218 align:middle line:84% the government will charge Ben and Jerry equal amounts. 00:41:32.218 --> 00:41:33.510 align:middle line:90% So there'll be equal financing. 00:41:33.510 --> 00:41:35.343 align:middle line:84% Just because they're individuals, identical, 00:41:35.343 --> 00:41:38.030 align:middle line:90% we'll have equal financing. 00:41:38.030 --> 00:41:43.430 align:middle line:84% Finally, we're going to assume the government provides 00:41:43.430 --> 00:41:47.970 align:middle line:84% fewer fireworks than were provided individually. 00:41:47.970 --> 00:41:50.430 align:middle line:84% The government is going to provide fewer fireworks 00:41:50.430 --> 00:41:52.190 align:middle line:90% than the private optimum-- 00:41:52.190 --> 00:41:54.540 align:middle line:90% less than 40. 00:41:54.540 --> 00:41:59.500 align:middle line:84% Under those three conditions, I will now show you mathematically 00:41:59.500 --> 00:42:03.460 align:middle line:84% that government provision does nothing. 00:42:03.460 --> 00:42:05.260 align:middle line:90% Let's see that mathematically. 00:42:05.260 --> 00:42:14.380 align:middle line:84% So under those conditions, remember that basically-- 00:42:14.380 --> 00:42:17.420 align:middle line:84% we said in the original example, total fireworks under-provided 00:42:17.420 --> 00:42:18.527 align:middle line:90% by 26.6. 00:42:18.527 --> 00:42:19.360 align:middle line:90% It's covered up now. 00:42:19.360 --> 00:42:22.980 align:middle line:84% But I said the social optimum is 66.6, private is 40. 00:42:22.980 --> 00:42:24.915 align:middle line:90% So 26.6 is the underprovision. 00:42:24.915 --> 00:42:26.540 align:middle line:84% So imagine the government says, well, I 00:42:26.540 --> 00:42:28.100 align:middle line:90% know how to solve this problem. 00:42:28.100 --> 00:42:31.340 align:middle line:84% I'm going to charge Ben and Jerry each 13.3 00:42:31.340 --> 00:42:33.420 align:middle line:90% and have public fireworks. 00:42:33.420 --> 00:42:35.740 align:middle line:84% So they'll provide their 40 privately. 00:42:35.740 --> 00:42:38.260 align:middle line:90% I'll provide my 26 extra. 00:42:38.260 --> 00:42:39.840 align:middle line:84% And I'll pay for it by charging them. 00:42:39.840 --> 00:42:43.500 align:middle line:90% I'll force them to contribute. 00:42:43.500 --> 00:42:48.460 align:middle line:84% So what I do is set up a tax on each of them of 13.33. 00:42:48.460 --> 00:42:53.730 align:middle line:84% Now their utility functions become u equals 2 times 00:42:53.730 --> 00:43:12.570 align:middle line:84% log x 2 times log x plus log fB plus fj plus 26.6, 00:43:12.570 --> 00:43:15.170 align:middle line:84% because they're getting 26.6 extra fireworks 00:43:15.170 --> 00:43:17.490 align:middle line:90% from the government. 00:43:17.490 --> 00:43:24.050 align:middle line:84% But subject to the budget constraint that x plus x sub i 00:43:24.050 --> 00:43:27.570 align:middle line:84% plus f sub i-- these are all i's. 00:43:27.570 --> 00:43:32.970 align:middle line:84% x sub i plus f sub i equals 100 minus 13.3 00:43:32.970 --> 00:43:34.390 align:middle line:90% So here's the new problem. 00:43:34.390 --> 00:43:36.243 align:middle line:90% We're taxing 13.3. 00:43:36.243 --> 00:43:37.910 align:middle line:84% That comes out of the budget constraint. 00:43:37.910 --> 00:43:42.090 align:middle line:84% And we're giving an extra 26.6 fireworks. 00:43:42.090 --> 00:43:44.930 align:middle line:90% People understand that setup? 00:43:44.930 --> 00:43:50.470 align:middle line:84% If you solve this, you will find that each individual chooses 00:43:50.470 --> 00:43:54.670 align:middle line:84% to provide 20, and the total remains at 40. 00:43:54.670 --> 00:43:57.910 align:middle line:84% That is, we will get what we call full crowd out, 00:43:57.910 --> 00:44:02.310 align:middle line:84% full displacement of the public expenditure 00:44:02.310 --> 00:44:04.970 align:middle line:84% by reduced private contributions. 00:44:04.970 --> 00:44:06.470 align:middle line:90% Why? 00:44:06.470 --> 00:44:09.970 align:middle line:90% 00:44:09.970 --> 00:44:12.930 align:middle line:84% Why will government intervention in this case make no difference? 00:44:12.930 --> 00:44:21.790 align:middle line:90% 00:44:21.790 --> 00:44:22.330 align:middle line:90% Yeah. 00:44:22.330 --> 00:44:27.710 align:middle line:84% AUDIENCE: [INAUDIBLE] Ben and Jerry are purchasing the 13.3, 00:44:27.710 --> 00:44:29.248 align:middle line:90% it's just, they're not directly. 00:44:29.248 --> 00:44:30.790 align:middle line:84% They're going through the government. 00:44:30.790 --> 00:44:33.480 align:middle line:84% So now they'll just subtract that off [INAUDIBLE]. 00:44:33.480 --> 00:44:34.230 align:middle line:90% JON GRUBER: Right. 00:44:34.230 --> 00:44:39.150 align:middle line:84% Remember, economic equilibria are fundamentally robust. 00:44:39.150 --> 00:44:42.530 align:middle line:84% And if the government intervenes in a way that can be undone, 00:44:42.530 --> 00:44:44.310 align:middle line:90% it will be undone. 00:44:44.310 --> 00:44:47.150 align:middle line:84% Government interventions only matter if they 00:44:47.150 --> 00:44:48.233 align:middle line:90% can't be perfectly undone. 00:44:48.233 --> 00:44:50.942 align:middle line:84% And literally, you can perfectly undo the government intervention 00:44:50.942 --> 00:44:51.500 align:middle line:90% here. 00:44:51.500 --> 00:44:53.160 align:middle line:84% There's no distortion, nothing else. 00:44:53.160 --> 00:44:55.880 align:middle line:84% You just literally say, fine, I'll provide 13.3 less myself. 00:44:55.880 --> 00:44:56.960 align:middle line:90% I'll be back where I was. 00:44:56.960 --> 00:44:58.860 align:middle line:84% And remember, that's where I want to be. 00:44:58.860 --> 00:45:01.260 align:middle line:90% I already showed you my optimum. 00:45:01.260 --> 00:45:03.220 align:middle line:84% I can get back to that same optimum 00:45:03.220 --> 00:45:08.340 align:middle line:84% by just lowering my contributions from 20 to 6.66. 00:45:08.340 --> 00:45:10.900 align:middle line:84% And boom, I'm back to the optimum. 00:45:10.900 --> 00:45:13.478 align:middle line:84% People, we already know where they're happiest. 00:45:13.478 --> 00:45:15.020 align:middle line:84% If the government intervention allows 00:45:15.020 --> 00:45:17.145 align:middle line:84% them to get exactly back to where they're happiest, 00:45:17.145 --> 00:45:18.340 align:middle line:90% they'll do so. 00:45:18.340 --> 00:45:19.800 align:middle line:84% So under these assumptions I made, 00:45:19.800 --> 00:45:23.180 align:middle line:90% there will be full crowd out. 00:45:23.180 --> 00:45:26.220 align:middle line:84% That is, government intervention will not actually 00:45:26.220 --> 00:45:30.420 align:middle line:90% increase the level of provision. 00:45:30.420 --> 00:45:33.127 align:middle line:90% Questions about that? 00:45:33.127 --> 00:45:33.960 align:middle line:90% Folks look confused. 00:45:33.960 --> 00:45:37.220 align:middle line:90% So please ask. 00:45:37.220 --> 00:45:38.000 align:middle line:90% Yeah, Enoch. 00:45:38.000 --> 00:45:40.740 align:middle line:84% AUDIENCE: For example, if you think about government donating 00:45:40.740 --> 00:45:43.740 align:middle line:84% to charity, like in the sense of we pay taxes, 00:45:43.740 --> 00:45:46.340 align:middle line:84% instead of us giving to charity, government gives to charity. 00:45:46.340 --> 00:45:46.840 align:middle line:90% Would that be an example of-- 00:45:46.840 --> 00:45:48.590 align:middle line:90% JON GRUBER: That's an example. 00:45:48.590 --> 00:45:52.650 align:middle line:84% Now, this full credit will almost never happen. 00:45:52.650 --> 00:45:56.370 align:middle line:84% In your example, why would it not lead to full crowd out? 00:45:56.370 --> 00:46:00.630 align:middle line:84% AUDIENCE: People still have utility from giving. 00:46:00.630 --> 00:46:03.730 align:middle line:84% JON GRUBER: Let's take that out of it. 00:46:03.730 --> 00:46:04.352 align:middle line:90% Yeah. 00:46:04.352 --> 00:46:05.810 align:middle line:84% AUDIENCE: The government's probably 00:46:05.810 --> 00:46:08.070 align:middle line:84% paying for the taxes for someone who wasn't going to donate. 00:46:08.070 --> 00:46:09.010 align:middle line:90% JON GRUBER: Exactly. 00:46:09.010 --> 00:46:11.850 align:middle line:84% The key thing is in my example, I 00:46:11.850 --> 00:46:14.370 align:middle line:84% set it up so the people were taxed in proportion 00:46:14.370 --> 00:46:16.650 align:middle line:90% to their contributions. 00:46:16.650 --> 00:46:20.010 align:middle line:84% Imagine a world where the government comes in 00:46:20.010 --> 00:46:23.970 align:middle line:84% and the government actually taxes some people who 00:46:23.970 --> 00:46:25.410 align:middle line:90% weren't contributing before. 00:46:25.410 --> 00:46:29.250 align:middle line:90% Well, then the total will go up. 00:46:29.250 --> 00:46:30.890 align:middle line:84% Remember, the key assumption I made 00:46:30.890 --> 00:46:34.010 align:middle line:84% is that each individual is taxed by less than what 00:46:34.010 --> 00:46:35.890 align:middle line:90% they were contributing before. 00:46:35.890 --> 00:46:38.430 align:middle line:84% Once everyone is taxed by more than contributing before, 00:46:38.430 --> 00:46:40.490 align:middle line:90% then you'll get more. 00:46:40.490 --> 00:46:45.070 align:middle line:84% So crowding out is unlikely to ever be full. 00:46:45.070 --> 00:46:47.350 align:middle line:84% Credit's unlikely to ever be full. 00:46:47.350 --> 00:46:49.870 align:middle line:84% But it almost certainly will always 00:46:49.870 --> 00:46:52.910 align:middle line:84% be partial because there will always 00:46:52.910 --> 00:46:55.650 align:middle line:84% be some view that, if the government is contributing, 00:46:55.650 --> 00:46:56.997 align:middle line:90% I can contribute less. 00:46:56.997 --> 00:46:58.830 align:middle line:84% And it's like my example, leaving the subway 00:46:58.830 --> 00:47:00.550 align:middle line:90% and seeing the homeless guy. 00:47:00.550 --> 00:47:02.510 align:middle line:84% You're being crowded out by the fact 00:47:02.510 --> 00:47:05.030 align:middle line:84% that other people are there giving. 00:47:05.030 --> 00:47:07.330 align:middle line:84% Now, there's a lot of studies of crowd out. 00:47:07.330 --> 00:47:09.550 align:middle line:84% Most of them are in lab settings. 00:47:09.550 --> 00:47:11.670 align:middle line:84% And they find very strong evidence 00:47:11.670 --> 00:47:14.510 align:middle line:90% for the existence of crowd out. 00:47:14.510 --> 00:47:19.830 align:middle line:84% They find very strong evidence for the existence of crowd out. 00:47:19.830 --> 00:47:23.450 align:middle line:90% Anyway, I can't find it. 00:47:23.450 --> 00:47:27.070 align:middle line:84% But there's a lot of examples of basically where they 00:47:27.070 --> 00:47:29.907 align:middle line:84% do laboratory studies where they essentially operationalize crowd 00:47:29.907 --> 00:47:31.990 align:middle line:84% out, where they basically have the government come 00:47:31.990 --> 00:47:34.190 align:middle line:84% in and substitute in for your contributions. 00:47:34.190 --> 00:47:38.390 align:middle line:84% In those lab studies, they find a lot of crowd out. 00:47:38.390 --> 00:47:41.150 align:middle line:84% In real-world empirical studies, the crowd out, people find, 00:47:41.150 --> 00:47:43.230 align:middle line:90% is much smaller. 00:47:43.230 --> 00:47:45.660 align:middle line:84% So I would say lab studies suggests that crowd out 00:47:45.660 --> 00:47:47.660 align:middle line:90% can be close to full. 00:47:47.660 --> 00:47:51.520 align:middle line:84% Real world says crowd out is less than 50% in many contexts. 00:47:51.520 --> 00:47:54.360 align:middle line:84% But it's hugely context-dependent, 00:47:54.360 --> 00:47:57.443 align:middle line:84% and this is a term we'll use a ton throughout the course. 00:47:57.443 --> 00:47:58.860 align:middle line:84% Basically, it's going to come back 00:47:58.860 --> 00:48:00.860 align:middle line:84% to crowd out, which is to what extent 00:48:00.860 --> 00:48:04.460 align:middle line:84% does government contribution displace private contribution? 00:48:04.460 --> 00:48:06.460 align:middle line:84% Here, it's a [INAUDIBLE] financial contribution. 00:48:06.460 --> 00:48:08.540 align:middle line:84% In health insurance, we'll talk about private health insurance 00:48:08.540 --> 00:48:10.140 align:middle line:84% versus a public health insurance. 00:48:10.140 --> 00:48:11.230 align:middle line:84% And unemployment insurance-- we'll 00:48:11.230 --> 00:48:13.605 align:middle line:84% talk about your savings versus the unemployment insurance 00:48:13.605 --> 00:48:14.113 align:middle line:90% benefit. 00:48:14.113 --> 00:48:15.280 align:middle line:90% But it's all the same thing. 00:48:15.280 --> 00:48:17.860 align:middle line:84% It's the notion that you can contribute privately. 00:48:17.860 --> 00:48:19.940 align:middle line:84% So the government rolls in and contributes. 00:48:19.940 --> 00:48:22.300 align:middle line:84% That will offset your private contributions. 00:48:22.300 --> 00:48:27.060 align:middle line:84% And the extent to which it will is an empirical question. 00:48:27.060 --> 00:48:27.560 align:middle line:90% Yeah. 00:48:27.560 --> 00:48:29.200 align:middle line:84% AUDIENCE: As long as the government's efficient, 00:48:29.200 --> 00:48:31.460 align:middle line:84% it doesn't matter if there's crowd out, necessarily? 00:48:31.460 --> 00:48:33.460 align:middle line:84% JON GRUBER: Well, that's a great question, which 00:48:33.460 --> 00:48:36.180 align:middle line:84% is what's the welfare consequence of crowd out? 00:48:36.180 --> 00:48:39.460 align:middle line:84% If the government has free money, 00:48:39.460 --> 00:48:41.440 align:middle line:84% then there's no welfare consequences. 00:48:41.440 --> 00:48:45.010 align:middle line:84% But in fact, money isn't free when money's raised by taxation. 00:48:45.010 --> 00:48:46.810 align:middle line:84% Now, in this case, it is free money 00:48:46.810 --> 00:48:49.710 align:middle line:84% because I don't have a distortion margin here. 00:48:49.710 --> 00:48:51.610 align:middle line:84% But if I had Ben and Jerry working 00:48:51.610 --> 00:48:53.965 align:middle line:84% and they could lower the tax burden by working less, 00:48:53.965 --> 00:48:56.090 align:middle line:84% that would be a distortion and make life worse than 00:48:56.090 --> 00:48:58.213 align:middle line:90% if the government didn't do it. 00:48:58.213 --> 00:49:00.130 align:middle line:84% AUDIENCE: But it could also be that they could 00:49:00.130 --> 00:49:01.310 align:middle line:90% produce fireworks cheaper. 00:49:01.310 --> 00:49:02.810 align:middle line:84% So the cost to the government [INAUDIBLE]. 00:49:02.810 --> 00:49:04.727 align:middle line:84% JON GRUBER: Right, so what if the government's 00:49:04.727 --> 00:49:06.750 align:middle line:90% cheaper producing them? 00:49:06.750 --> 00:49:08.015 align:middle line:90% So you're absolutely right. 00:49:08.015 --> 00:49:10.390 align:middle line:84% In the real world, that would depend on a lot of factors. 00:49:10.390 --> 00:49:11.530 align:middle line:84% One is the crowd out would depend 00:49:11.530 --> 00:49:12.650 align:middle line:90% on the efficiency of producing. 00:49:12.650 --> 00:49:14.490 align:middle line:84% Like maybe the government is better producing health insurers 00:49:14.490 --> 00:49:15.650 align:middle line:90% than the private sector. 00:49:15.650 --> 00:49:18.250 align:middle line:84% So even if it crowds it out, that's still a good thing. 00:49:18.250 --> 00:49:20.633 align:middle line:84% It will also depend a lot on how bad crowd out is. 00:49:20.633 --> 00:49:23.050 align:middle line:84% Sometimes we crowd out things that people are doing, which 00:49:23.050 --> 00:49:25.130 align:middle line:90% we wish they weren't doing. 00:49:25.130 --> 00:49:28.698 align:middle line:84% So for example, talking about unemployment insurance, 00:49:28.698 --> 00:49:30.490 align:middle line:84% one way people react to being unemployed is 00:49:30.490 --> 00:49:33.450 align:middle line:84% by their spouses going to work, even if the spouses would 00:49:33.450 --> 00:49:34.255 align:middle line:90% prefer not to work. 00:49:34.255 --> 00:49:36.130 align:middle line:84% Well, in that case, if unemployment insurance 00:49:36.130 --> 00:49:38.790 align:middle line:84% allows the spouse to stay home and take care of the kids, 00:49:38.790 --> 00:49:40.270 align:middle line:84% maybe that's not such a bad thing. 00:49:40.270 --> 00:49:41.530 align:middle line:90% But we call it crowd out. 00:49:41.530 --> 00:49:42.030 align:middle line:90% So 00:49:42.030 --> 00:49:44.710 align:middle line:84% The welfare consequences crowd are very much dependent 00:49:44.710 --> 00:49:46.190 align:middle line:84% on the efficiency of the government 00:49:46.190 --> 00:49:47.690 align:middle line:90% versus the private provision. 00:49:47.690 --> 00:49:48.610 align:middle line:90% It's a great point. 00:49:48.610 --> 00:49:50.735 align:middle line:84% In this case, I assume they were equally efficient, 00:49:50.735 --> 00:49:53.950 align:middle line:84% so there's not really any welfare consequences. 00:49:53.950 --> 00:49:54.810 align:middle line:90% Great question. 00:49:54.810 --> 00:49:57.070 align:middle line:90% Other questions? 00:49:57.070 --> 00:50:01.910 align:middle line:84% So that's the first problem that we have. 00:50:01.910 --> 00:50:04.390 align:middle line:90% The first problem is crowd out. 00:50:04.390 --> 00:50:07.310 align:middle line:84% The second problem with public provision 00:50:07.310 --> 00:50:09.530 align:middle line:90% is the problem of measurement. 00:50:09.530 --> 00:50:16.310 align:middle line:90% 00:50:16.310 --> 00:50:20.310 align:middle line:84% I've written down the Samuelson formula, which is the sum of MRS 00:50:20.310 --> 00:50:21.410 align:middle line:90% equals the MRT. 00:50:21.410 --> 00:50:23.950 align:middle line:90% 00:50:23.950 --> 00:50:25.750 align:middle line:84% Well, that does you no good if you're 00:50:25.750 --> 00:50:28.510 align:middle line:84% trying to decide whether to build a bridge or not. 00:50:28.510 --> 00:50:29.590 align:middle line:90% We don't know people-- 00:50:29.590 --> 00:50:31.390 align:middle line:90% people don't have MRSs. 00:50:31.390 --> 00:50:34.050 align:middle line:84% And we sort of have MRTs, we know relative marginal costs. 00:50:34.050 --> 00:50:36.230 align:middle line:90% We certainly don't have MRSs. 00:50:36.230 --> 00:50:39.180 align:middle line:84% So we have to figure out how to actually measure 00:50:39.180 --> 00:50:42.060 align:middle line:90% costs and benefits in reality. 00:50:42.060 --> 00:50:44.200 align:middle line:84% We can't just use-- that's the theoretical concept. 00:50:44.200 --> 00:50:45.950 align:middle line:84% And that leads to the entire field of what 00:50:45.950 --> 00:50:47.840 align:middle line:90% we call cost-benefit analysis. 00:50:47.840 --> 00:50:51.780 align:middle line:90% 00:50:51.780 --> 00:50:53.820 align:middle line:84% We will discuss this next time in chapter 8, 00:50:53.820 --> 00:50:55.140 align:middle line:90% or we may start today. 00:50:55.140 --> 00:50:57.220 align:middle line:84% This is probably the most single thing 00:50:57.220 --> 00:50:59.800 align:middle line:84% that-- if you look at all economists around the world, 00:50:59.800 --> 00:51:01.060 align:middle line:84% in the real world, probably the single thing 00:51:01.060 --> 00:51:02.580 align:middle line:84% they spend the most time on is doing 00:51:02.580 --> 00:51:05.140 align:middle line:84% some or another version of cost-benefit analysis, which 00:51:05.140 --> 00:51:08.900 align:middle line:84% is actually quantifying costs and benefits that will lead 00:51:08.900 --> 00:51:11.540 align:middle line:84% to-- so you can actually decide if it's a decision worth making 00:51:11.540 --> 00:51:12.460 align:middle line:90% or not. 00:51:12.460 --> 00:51:14.660 align:middle line:84% So chapter 8 will be about how do we literally 00:51:14.660 --> 00:51:18.700 align:middle line:84% translate this theoretical construct into real measurement. 00:51:18.700 --> 00:51:22.900 align:middle line:84% The third problem is, let's say we know what's optimal. 00:51:22.900 --> 00:51:24.860 align:middle line:84% We know how much crowd out there is. 00:51:24.860 --> 00:51:26.580 align:middle line:90% We know how to measure it. 00:51:26.580 --> 00:51:28.580 align:middle line:84% We know how to measure the cost of benefits. 00:51:28.580 --> 00:51:31.520 align:middle line:84% The question is, how do we measure preferences? 00:51:31.520 --> 00:51:35.540 align:middle line:90% 00:51:35.540 --> 00:51:38.390 align:middle line:84% And here, in other words, how do we-- 00:51:38.390 --> 00:51:40.870 align:middle line:84% we're a society of 360 million people. 00:51:40.870 --> 00:51:44.250 align:middle line:84% How do we decide whether or not to do a public good? 00:51:44.250 --> 00:51:47.490 align:middle line:84% Because in fact, we don't have just one dictatorial public 00:51:47.490 --> 00:51:48.030 align:middle line:90% planner. 00:51:48.030 --> 00:51:50.130 align:middle line:90% We have a democracy. 00:51:50.130 --> 00:51:54.450 align:middle line:84% And that raises three critical problems. 00:51:54.450 --> 00:51:59.490 align:middle line:84% The first problem is preference revelation, 00:51:59.490 --> 00:52:02.170 align:middle line:84% which is we need to know people's preferences. 00:52:02.170 --> 00:52:06.570 align:middle line:84% And if we ask us to tell them, they may not be honest. 00:52:06.570 --> 00:52:10.370 align:middle line:84% The second problem is the problem 00:52:10.370 --> 00:52:15.290 align:middle line:84% of preference knowledge, which is people may not even 00:52:15.290 --> 00:52:16.967 align:middle line:90% know their preferences. 00:52:16.967 --> 00:52:19.050 align:middle line:84% I mean, do you know what a missile is worth to you 00:52:19.050 --> 00:52:19.883 align:middle line:90% in terms of cookies? 00:52:19.883 --> 00:52:22.650 align:middle line:84% You have no fucking idea, so you might not even 00:52:22.650 --> 00:52:23.670 align:middle line:90% know your preferences. 00:52:23.670 --> 00:52:25.130 align:middle line:90% He's saying seven cookies. 00:52:25.130 --> 00:52:31.090 align:middle line:84% OK, you may not even know your preferences, so this knowledge. 00:52:31.090 --> 00:52:34.390 align:middle line:84% The final problem is aggregation. 00:52:34.390 --> 00:52:37.005 align:middle line:90% 00:52:37.005 --> 00:52:41.170 align:middle line:84% Let's say that you know your preferences and you're honest. 00:52:41.170 --> 00:52:43.830 align:middle line:84% How do I collect all those preferences 00:52:43.830 --> 00:52:47.220 align:middle line:84% in a way that effectively aggregates them? 00:52:47.220 --> 00:52:49.470 align:middle line:84% That turns out to be hard to do for reasons we'll talk 00:52:49.470 --> 00:52:50.930 align:middle line:90% about when we do chapter 9. 00:52:50.930 --> 00:52:55.750 align:middle line:84% This all leads to the field of political economy in chapter 9-- 00:52:55.750 --> 00:52:59.910 align:middle line:90% political econ in chapter 9. 00:52:59.910 --> 00:53:03.870 align:middle line:90% So this is really the key. 00:53:03.870 --> 00:53:06.830 align:middle line:84% The key insights of chapter 7 are the following. 00:53:06.830 --> 00:53:10.750 align:middle line:84% A-- optimal provision of public goods 00:53:10.750 --> 00:53:14.350 align:middle line:84% is where the sum of the MRS equals the MRT. 00:53:14.350 --> 00:53:17.310 align:middle line:84% B-- the private sector is unlikely to achieve 00:53:17.310 --> 00:53:23.470 align:middle line:84% that optimum due to the free rider problem. 00:53:23.470 --> 00:53:26.250 align:middle line:84% C-- or was it A, B, C or 1, 2, 3? 00:53:26.250 --> 00:53:29.470 align:middle line:84% Whatever the third thing is on my list. 00:53:29.470 --> 00:53:33.150 align:middle line:84% The public sector itself faces challenges also 00:53:33.150 --> 00:53:35.720 align:middle line:90% in solving this problem. 00:53:35.720 --> 00:53:38.797 align:middle line:84% And those challenges are what we'll turn to next. 00:53:38.797 --> 00:53:40.880 align:middle line:84% Now, I want to cover one last thing for chapter 7. 00:53:40.880 --> 00:53:42.463 align:middle line:84% I won't give it as much time because I 00:53:42.463 --> 00:53:43.580 align:middle line:90% want to get to chapter 8. 00:53:43.580 --> 00:53:46.060 align:middle line:84% But there's an awesome box in chapter 7 00:53:46.060 --> 00:53:48.900 align:middle line:84% about the fundamental challenge the government faces 00:53:48.900 --> 00:53:52.080 align:middle line:84% in deciding whether to provide things publicly or privately. 00:53:52.080 --> 00:53:56.340 align:middle line:84% And this is the problem we call contracting out, which is, 00:53:56.340 --> 00:54:00.020 align:middle line:84% let's say the government knows there's too little of something. 00:54:00.020 --> 00:54:03.060 align:middle line:84% There's fundamentally two ways it can address that. 00:54:03.060 --> 00:54:07.940 align:middle line:84% It can address that by providing it or by-- 00:54:07.940 --> 00:54:10.720 align:middle line:84% let's say the government wants there to be more of something. 00:54:10.720 --> 00:54:12.887 align:middle line:84% There's two ways it can-- and not the only two ways, 00:54:12.887 --> 00:54:15.380 align:middle line:84% but two ways it can address that is by providing it 00:54:15.380 --> 00:54:18.005 align:middle line:84% or by hiring other private parties to provide it. 00:54:18.005 --> 00:54:19.380 align:middle line:84% So should the government directly 00:54:19.380 --> 00:54:22.340 align:middle line:84% provide something or contract out? 00:54:22.340 --> 00:54:24.540 align:middle line:84% And it turns out contracting out raises 00:54:24.540 --> 00:54:27.300 align:middle line:84% a host of fascinating issues around how do you 00:54:27.300 --> 00:54:30.140 align:middle line:84% write private contracts to get the outcome you want? 00:54:30.140 --> 00:54:33.587 align:middle line:84% And there's great examples of all sorts of shenanigans 00:54:33.587 --> 00:54:34.670 align:middle line:90% when the government tries. 00:54:34.670 --> 00:54:36.310 align:middle line:84% There's an example of good contracting out, 00:54:36.310 --> 00:54:37.643 align:middle line:90% but that's no fun to talk about. 00:54:37.643 --> 00:54:39.370 align:middle line:84% Let's talk about the shenanigans. 00:54:39.370 --> 00:54:43.290 align:middle line:84% The shenanigans are that basically, what we 00:54:43.290 --> 00:54:44.890 align:middle line:84% find is a number of studies which 00:54:44.890 --> 00:54:50.370 align:middle line:84% show that when the government contracts out social services, 00:54:50.370 --> 00:54:54.570 align:middle line:90% it gets low-quality outcomes. 00:54:54.570 --> 00:54:56.550 align:middle line:84% When it contracts out ambulances, 00:54:56.550 --> 00:55:00.450 align:middle line:84% prisons-- private prisons have 10% more deaths 00:55:00.450 --> 00:55:02.210 align:middle line:90% than public prisons. 00:55:02.210 --> 00:55:05.030 align:middle line:84% Halfway houses-- horror stories about halfway houses, 00:55:05.030 --> 00:55:09.443 align:middle line:84% which are places people go when they're trying to get clean. 00:55:09.443 --> 00:55:11.610 align:middle line:84% There's horrible stories of halfway houses being run 00:55:11.610 --> 00:55:14.930 align:middle line:84% by for-profit private enterprises that basically, 00:55:14.930 --> 00:55:15.830 align:middle line:90% the government-- 00:55:15.830 --> 00:55:18.770 align:middle line:84% Now, on the other hand, there's a long, long literature 00:55:18.770 --> 00:55:21.955 align:middle line:84% that privatization of many sectors, 00:55:21.955 --> 00:55:23.330 align:middle line:84% that the government traditionally 00:55:23.330 --> 00:55:25.650 align:middle line:90% provided back in the day-- 00:55:25.650 --> 00:55:28.130 align:middle line:84% and in many places in the world, governments often 00:55:28.130 --> 00:55:33.010 align:middle line:84% provided things like steel, airplanes, banks. 00:55:33.010 --> 00:55:34.275 align:middle line:90% Those were all public. 00:55:34.275 --> 00:55:35.650 align:middle line:84% Those were all publicly provided. 00:55:35.650 --> 00:55:37.130 align:middle line:84% And in many countries like the US, 00:55:37.130 --> 00:55:38.770 align:middle line:90% those became privately provided. 00:55:38.770 --> 00:55:41.470 align:middle line:84% There's a lot of evidence that when you're providing goods, 00:55:41.470 --> 00:55:44.710 align:middle line:84% there's efficiencies from the private sector taking it over. 00:55:44.710 --> 00:55:46.870 align:middle line:84% But when you're providing services, 00:55:46.870 --> 00:55:49.175 align:middle line:84% they get worse when the private sector takes it over. 00:55:49.175 --> 00:55:50.550 align:middle line:84% And the fundamental reason is one 00:55:50.550 --> 00:55:52.430 align:middle line:84% Coase would have us focus on, which 00:55:52.430 --> 00:55:56.448 align:middle line:84% is essentially observability of quality, which 00:55:56.448 --> 00:55:58.990 align:middle line:84% is, I can really observe easily if it's a good piece of steel 00:55:58.990 --> 00:55:59.730 align:middle line:90% or not. 00:55:59.730 --> 00:56:02.950 align:middle line:84% It's hard to observe it's a good prison or good halfway house. 00:56:02.950 --> 00:56:04.770 align:middle line:84% So when you have less observable quality, 00:56:04.770 --> 00:56:07.750 align:middle line:84% which you have with services, the private sector 00:56:07.750 --> 00:56:09.870 align:middle line:90% contracting out can have costs. 00:56:09.870 --> 00:56:13.310 align:middle line:84% Now, they can try to combat those costs with monitoring. 00:56:13.310 --> 00:56:19.670 align:middle line:84% So for example, Rikers Island jail facility in New York City 00:56:19.670 --> 00:56:23.290 align:middle line:90% had a spate of violence in 2014. 00:56:23.290 --> 00:56:25.970 align:middle line:90% So they paid McKinsey Company-- 00:56:25.970 --> 00:56:29.110 align:middle line:84% they're bad guys in all these stories, $27.5 million 00:56:29.110 --> 00:56:31.980 align:middle line:84% to create a plan to curtail violence in the faculty. 00:56:31.980 --> 00:56:33.780 align:middle line:84% Three years later, McKinsey reported 00:56:33.780 --> 00:56:36.500 align:middle line:84% that their $27.5 million plan had succeeded 00:56:36.500 --> 00:56:42.340 align:middle line:84% and that violence in the facility had fallen by 70%. 00:56:42.340 --> 00:56:45.460 align:middle line:84% It turns out McKinsey rigged the results and violence went up 00:56:45.460 --> 00:56:47.460 align:middle line:90% by 50% 00:56:47.460 --> 00:56:48.920 align:middle line:90% That's not even the worst case. 00:56:48.920 --> 00:56:50.920 align:middle line:84% The worst case involves one of my favorite names 00:56:50.920 --> 00:56:53.120 align:middle line:84% for a company, the Wackenhut Corporation. 00:56:53.120 --> 00:56:55.340 align:middle line:84% The Wackenhut Corporation is in charge 00:56:55.340 --> 00:56:57.700 align:middle line:84% of ensuring that our nuclear facilities are 00:56:57.700 --> 00:57:00.060 align:middle line:90% safe from military attack. 00:57:00.060 --> 00:57:04.305 align:middle line:84% So they are in charge of essentially probing and testing 00:57:04.305 --> 00:57:06.680 align:middle line:84% whether nuclear facilities are safe from military attack. 00:57:06.680 --> 00:57:09.860 align:middle line:84% And they're supposed to do this with simulated attacks. 00:57:09.860 --> 00:57:12.060 align:middle line:84% It turned out that the Wackenhut Corporation 00:57:12.060 --> 00:57:14.000 align:middle line:84% would warn these nuclear facilities 00:57:14.000 --> 00:57:17.060 align:middle line:84% that the attacks were coming and get paid for those warnings. 00:57:17.060 --> 00:57:20.100 align:middle line:84% And they all looked super-safe when they, in fact, weren't. 00:57:20.100 --> 00:57:25.040 align:middle line:84% So the problem is, when you're trying to contract out, 00:57:25.040 --> 00:57:26.880 align:middle line:90% you get potential efficiencies. 00:57:26.880 --> 00:57:30.850 align:middle line:84% But the monitoring difficulties can be very, very high. 00:57:30.850 --> 00:57:33.850 align:middle line:84% And that's a real challenge in of contracting out world. 00:57:33.850 --> 00:57:35.350 align:middle line:84% Anyway, it's a long box in the book, 00:57:35.350 --> 00:57:36.310 align:middle line:90% but it's kind of fun to read. 00:57:36.310 --> 00:57:37.727 align:middle line:84% And there's a lot of fun examples. 00:57:37.727 --> 00:57:41.490 align:middle line:84% And in some sense, sometimes the government has no choice, 00:57:41.490 --> 00:57:44.410 align:middle line:84% or faced difficulties, which is like during COVID. 00:57:44.410 --> 00:57:46.450 align:middle line:84% During COVID, we suddenly needed a shitload 00:57:46.450 --> 00:57:49.130 align:middle line:90% of masks and respirators. 00:57:49.130 --> 00:57:50.630 align:middle line:90% So we contracted out. 00:57:50.630 --> 00:57:51.390 align:middle line:90% We got a lot. 00:57:51.390 --> 00:57:53.610 align:middle line:90% We got a ton of fraud. 00:57:53.610 --> 00:57:55.910 align:middle line:84% A ton of companies, which weren't really capable, 00:57:55.910 --> 00:57:58.170 align:middle line:84% got big contracts and bombed out. 00:57:58.170 --> 00:58:00.890 align:middle line:84% We also got a lot of masks and respirators. 00:58:00.890 --> 00:58:03.430 align:middle line:84% That's why a lot of people were in favor of, 00:58:03.430 --> 00:58:05.930 align:middle line:84% instead of contracting that out, actually invoking something 00:58:05.930 --> 00:58:08.930 align:middle line:84% called the Defense Production Act, which 00:58:08.930 --> 00:58:11.130 align:middle line:84% grants the president authority to literally tell 00:58:11.130 --> 00:58:12.730 align:middle line:90% firms what to do. 00:58:12.730 --> 00:58:14.490 align:middle line:90% That's scary. 00:58:14.490 --> 00:58:19.190 align:middle line:84% But the idea is, if you do that, then you get the firms to do it. 00:58:19.190 --> 00:58:22.610 align:middle line:84% If you contract out, you might get firms that don't exist. 00:58:22.610 --> 00:58:25.747 align:middle line:84% Or when they're trying to cover Hurricane Maria in Puerto Rico, 00:58:25.747 --> 00:58:27.830 align:middle line:84% they hired some firm which turned out to be, like, 00:58:27.830 --> 00:58:30.030 align:middle line:84% two guys in Montana who happened to be buddies 00:58:30.030 --> 00:58:33.510 align:middle line:84% of the president at the time, who ended up not doing anything 00:58:33.510 --> 00:58:34.450 align:middle line:90% for Hurricane Maria. 00:58:34.450 --> 00:58:36.630 align:middle line:84% So there's an interesting trade-off here 00:58:36.630 --> 00:58:38.160 align:middle line:84% that is worth thinking more about. 00:58:38.160 --> 00:58:39.910 align:middle line:84% I wish I had more time on that, but that's 00:58:39.910 --> 00:58:43.270 align:middle line:90% all I can say about that. 00:58:43.270 --> 00:58:44.250 align:middle line:90% So that's chapter 7. 00:58:44.250 --> 00:58:46.590 align:middle line:90% Any other questions about that? 00:58:46.590 --> 00:58:49.350 align:middle line:90% OK, let's go on to chapter 8. 00:58:49.350 --> 00:58:52.590 align:middle line:84% And let's start thinking about this second problem-- 00:58:52.590 --> 00:58:57.190 align:middle line:84% measurement and the issue of cost-benefit analysis. 00:58:57.190 --> 00:58:59.130 align:middle line:84% And we're going to do this through an example. 00:58:59.130 --> 00:59:00.070 align:middle line:84% We're going have an example that's going 00:59:00.070 --> 00:59:01.010 align:middle line:90% to run through this chapter. 00:59:01.010 --> 00:59:01.650 align:middle line:90% We'll start it today. 00:59:01.650 --> 00:59:02.890 align:middle line:90% We'll finish it next time. 00:59:02.890 --> 00:59:05.630 align:middle line:84% [? An ?] example is going to run through this chapter 00:59:05.630 --> 00:59:09.630 align:middle line:84% of a government's decision to build a highway. 00:59:09.630 --> 00:59:13.790 align:middle line:84% And we're going to assume that building this highway, 00:59:13.790 --> 00:59:18.470 align:middle line:84% if you look at table 8-1, will require the following inputs-- 00:59:18.470 --> 00:59:23.350 align:middle line:84% a million bags of asphalt, million hours of labor, 00:59:23.350 --> 00:59:27.020 align:middle line:84% and $10 million a year of maintenance. 00:59:27.020 --> 00:59:30.240 align:middle line:84% But by building this-- let's say it's expanding a highway. 00:59:30.240 --> 00:59:35.580 align:middle line:84% Through this project, we'll save 500,000 hours per year 00:59:35.580 --> 00:59:38.180 align:middle line:84% in driving time because we have better highway. 00:59:38.180 --> 00:59:41.940 align:middle line:84% And it will save five lives a year because it'll be safer. 00:59:41.940 --> 00:59:45.420 align:middle line:90% So we have here the costs. 00:59:45.420 --> 00:59:47.785 align:middle line:90% We have the benefits. 00:59:47.785 --> 00:59:50.160 align:middle line:84% We know we want to sum those benefits equal to the costs. 00:59:50.160 --> 00:59:51.040 align:middle line:90% But we just have words. 00:59:51.040 --> 00:59:51.960 align:middle line:90% We don't have numbers. 00:59:51.960 --> 00:59:52.840 align:middle line:90% This is MIT. 00:59:52.840 --> 00:59:54.420 align:middle line:90% We need numbers. 00:59:54.420 --> 00:59:56.980 align:middle line:84% So how do we translate these words into numbers? 00:59:56.980 --> 01:00:01.420 align:middle line:84% And that is the field of cost-benefit analysis. 01:00:01.420 --> 01:00:06.420 align:middle line:84% Let's start with the easier side of this, which is the cost side. 01:00:06.420 --> 01:00:10.260 align:middle line:84% And let's think about how we measure the costs of things 01:00:10.260 --> 01:00:12.740 align:middle line:90% like asphalt and labor. 01:00:12.740 --> 01:00:16.140 align:middle line:84% And here, we're going to rag on our friends the accountants. 01:00:16.140 --> 01:00:19.500 align:middle line:84% The way accountants would do this 01:00:19.500 --> 01:00:21.620 align:middle line:84% is they would say, well, whatever 01:00:21.620 --> 01:00:25.170 align:middle line:84% you pay for the labor of the asphalt is its cost. 01:00:25.170 --> 01:00:26.970 align:middle line:84% And economists would say that's not true, 01:00:26.970 --> 01:00:30.590 align:middle line:84% because what matters is not the cash costs. 01:00:30.590 --> 01:00:33.530 align:middle line:90% It's the opportunity costs. 01:00:33.530 --> 01:00:37.810 align:middle line:84% It's the best alternative use of those resources in society. 01:00:37.810 --> 01:00:41.410 align:middle line:90% And those are not always equal. 01:00:41.410 --> 01:00:43.810 align:middle line:84% In a perfectly competitive market, 01:00:43.810 --> 01:00:46.090 align:middle line:90% they are by definition equal. 01:00:46.090 --> 01:00:49.615 align:middle line:84% In a perfectly competitive market, the price of a good 01:00:49.615 --> 01:00:50.990 align:middle line:84% is equal to its opportunity cost. 01:00:50.990 --> 01:00:53.282 align:middle line:84% Think about our perfectly competitive market and demand 01:00:53.282 --> 01:00:54.690 align:middle line:90% and supply equilibrating. 01:00:54.690 --> 01:00:57.770 align:middle line:84% They equilibrate the point where the value of the next bag 01:00:57.770 --> 01:01:00.130 align:middle line:84% of asphalt equals the cost of producing 01:01:00.130 --> 01:01:04.390 align:middle line:84% the next bag of asphalt. So the opportunity cost, 01:01:04.390 --> 01:01:06.370 align:middle line:84% which is producing one less bag of asphalt, 01:01:06.370 --> 01:01:09.370 align:middle line:84% is equal to the price, so cash accounting works. 01:01:09.370 --> 01:01:12.570 align:middle line:84% But now imagine the market is imperfect. 01:01:12.570 --> 01:01:16.010 align:middle line:84% And let's actually think about the labor market to start. 01:01:16.010 --> 01:01:20.410 align:middle line:84% Imagine that you have a world where 01:01:20.410 --> 01:01:24.310 align:middle line:84% construction workers, as they do in many cities in America, 01:01:24.310 --> 01:01:28.270 align:middle line:84% have a higher minimum wage for public projects 01:01:28.270 --> 01:01:31.630 align:middle line:90% than the statewide minimum wage. 01:01:31.630 --> 01:01:36.870 align:middle line:84% So let's imagine the statewide market wage is $10 an hour, 01:01:36.870 --> 01:01:39.590 align:middle line:84% that the state on average people earn $10 an hour. 01:01:39.590 --> 01:01:42.350 align:middle line:84% But there's a minimum wage for public construction projects 01:01:42.350 --> 01:01:43.570 align:middle line:90% of $20 an hour. 01:01:43.570 --> 01:01:46.550 align:middle line:90% 01:01:46.550 --> 01:01:52.270 align:middle line:84% So let's imagine that the government is going to hire 01:01:52.270 --> 01:01:58.630 align:middle line:90% these 500,000 workers. 01:01:58.630 --> 01:02:02.910 align:middle line:84% And it's going to end up paying $20 million, 01:02:02.910 --> 01:02:09.070 align:middle line:84% because it's $20 an hour for this million hours of labor. 01:02:09.070 --> 01:02:14.430 align:middle line:84% What is the opportunity cost of that labor? 01:02:14.430 --> 01:02:17.670 align:middle line:84% Well, it's the next best use of that labor. 01:02:17.670 --> 01:02:20.740 align:middle line:84% And the next best use is another project where 01:02:20.740 --> 01:02:22.860 align:middle line:90% they'd be paid $10 an hour. 01:02:22.860 --> 01:02:24.560 align:middle line:84% And assuming labor markets are perfect, 01:02:24.560 --> 01:02:26.442 align:middle line:90% that's their marginal product. 01:02:26.442 --> 01:02:27.900 align:middle line:84% Their marginal product is what they 01:02:27.900 --> 01:02:31.940 align:middle line:84% get paid in an alternative job, which is $10 an hour. 01:02:31.940 --> 01:02:37.060 align:middle line:84% So while the cash cost of these workers is $20 million, 01:02:37.060 --> 01:02:39.700 align:middle line:84% the opportunity cost is only $10 million. 01:02:39.700 --> 01:02:41.820 align:middle line:84% You're only extracting $10 million 01:02:41.820 --> 01:02:44.003 align:middle line:84% worth of marginal product out of society, 01:02:44.003 --> 01:02:45.920 align:middle line:84% because that's what they'd earn in other jobs, 01:02:45.920 --> 01:02:48.620 align:middle line:84% even though you're paying them $20 million. 01:02:48.620 --> 01:02:49.800 align:middle line:90% Well, then what's the magic? 01:02:49.800 --> 01:02:50.860 align:middle line:84% Where's the other $10 million go? 01:02:50.860 --> 01:02:52.277 align:middle line:84% I mean, the accountants are right, 01:02:52.277 --> 01:02:53.820 align:middle line:90% we're spending $20 million. 01:02:53.820 --> 01:02:56.260 align:middle line:84% What happens to the other $10 million? 01:02:56.260 --> 01:02:59.940 align:middle line:84% If the cash cost is $20 million, but the opportunity cost 01:02:59.940 --> 01:03:03.040 align:middle line:84% in terms of the marginal product we're using is $10 million, 01:03:03.040 --> 01:03:04.320 align:middle line:90% what's the other $10 million? 01:03:04.320 --> 01:03:08.040 align:middle line:90% 01:03:08.040 --> 01:03:08.540 align:middle line:90% Yeah, 01:03:08.540 --> 01:03:10.035 align:middle line:90% AUDIENCE: [INAUDIBLE] workers. 01:03:10.035 --> 01:03:11.660 align:middle line:84% JON GRUBER: Yeah, the other $10 million 01:03:11.660 --> 01:03:14.100 align:middle line:90% is what we call economic rents. 01:03:14.100 --> 01:03:16.980 align:middle line:84% Those workers, by getting these jobs, 01:03:16.980 --> 01:03:20.690 align:middle line:84% have gotten a transfer from the government. 01:03:20.690 --> 01:03:24.010 align:middle line:84% The government has transferred $10 an hour 01:03:24.010 --> 01:03:26.783 align:middle line:84% to workers by giving them these better jobs. 01:03:26.783 --> 01:03:28.950 align:middle line:84% It's not better in terms of higher marginal product, 01:03:28.950 --> 01:03:31.250 align:middle line:90% better in terms of higher pay. 01:03:31.250 --> 01:03:35.130 align:middle line:90% That has no efficiency cost. 01:03:35.130 --> 01:03:37.410 align:middle line:90% If I give you $1-- 01:03:37.410 --> 01:03:38.970 align:middle line:84% I just give you the dollar, there's 01:03:38.970 --> 01:03:41.610 align:middle line:90% no efficiency cost to that. 01:03:41.610 --> 01:03:46.050 align:middle line:84% So there's the opportunity cost, the efficiency cost to society, 01:03:46.050 --> 01:03:48.930 align:middle line:84% is the lost marginal product by building this road. 01:03:48.930 --> 01:03:50.853 align:middle line:84% It's what the workers could have done instead. 01:03:50.853 --> 01:03:52.270 align:middle line:84% If they weren't building the road, 01:03:52.270 --> 01:03:53.050 align:middle line:84% they would have been doing something 01:03:53.050 --> 01:03:55.210 align:middle line:84% with the marginal product of $10 an hour. 01:03:55.210 --> 01:03:56.430 align:middle line:90% And how do we know that? 01:03:56.430 --> 01:03:58.570 align:middle line:84% Because their wage is $10 an hour, and wage 01:03:58.570 --> 01:04:01.650 align:middle line:84% equals marginal product in a labor market equilibrium. 01:04:01.650 --> 01:04:03.530 align:middle line:90% Once again, let me say it again. 01:04:03.530 --> 01:04:06.210 align:middle line:84% In a labor market equilibrium, in a perfectly functioning labor 01:04:06.210 --> 01:04:09.070 align:middle line:84% market, wage equals marginal product. 01:04:09.070 --> 01:04:11.930 align:middle line:90% We solved for that in 14.01. 01:04:11.930 --> 01:04:15.210 align:middle line:84% Therefore the value of their time 01:04:15.210 --> 01:04:17.750 align:middle line:84% is the $10 marginal product we're losing 01:04:17.750 --> 01:04:19.163 align:middle line:90% by having them build this road. 01:04:19.163 --> 01:04:20.330 align:middle line:90% That's the opportunity cost. 01:04:20.330 --> 01:04:22.670 align:middle line:84% The other $10 million is literally just a transfer 01:04:22.670 --> 01:04:24.628 align:middle line:84% from the government to the lucky set of workers 01:04:24.628 --> 01:04:26.710 align:middle line:90% who happen to get these jobs. 01:04:26.710 --> 01:04:28.690 align:middle line:90% It's a hard concept. 01:04:28.690 --> 01:04:30.190 align:middle line:90% Questions about that? 01:04:30.190 --> 01:04:30.790 align:middle line:90% Yeah. 01:04:30.790 --> 01:04:32.630 align:middle line:84% AUDIENCE: How is that margin [INAUDIBLE]? 01:04:32.630 --> 01:04:33.810 align:middle line:90% JON GRUBER: How's what? 01:04:33.810 --> 01:04:37.310 align:middle line:84% AUDIENCE: The margin between $20 and $10 set by the government 01:04:37.310 --> 01:04:39.132 align:middle line:90% outside of, maybe, [INAUDIBLE]? 01:04:39.132 --> 01:04:41.590 align:middle line:84% JON GRUBER: You mean where does that arise from in reality? 01:04:41.590 --> 01:04:43.070 align:middle line:90% Just political economy. 01:04:43.070 --> 01:04:45.630 align:middle line:84% They just decide they're going to-- the construction workers 01:04:45.630 --> 01:04:47.390 align:middle line:90% lobby, the union lobbies. 01:04:47.390 --> 01:04:48.890 align:middle line:84% And they get a higher wage for that. 01:04:48.890 --> 01:04:51.653 align:middle line:90% 01:04:51.653 --> 01:04:53.070 align:middle line:84% That leads us to this point, which 01:04:53.070 --> 01:04:55.170 align:middle line:90% is why do we do what we do? 01:04:55.170 --> 01:04:58.030 align:middle line:90% 01:04:58.030 --> 01:05:03.230 align:middle line:84% Now, can anyone tell me why this same logic might apply, 01:05:03.230 --> 01:05:06.350 align:middle line:84% and why the opportunity cost, or the efficiency cost 01:05:06.350 --> 01:05:10.630 align:middle line:84% of asphalt, under what condition might the opportunity 01:05:10.630 --> 01:05:14.050 align:middle line:84% cost of asphalt be below its cash cost? 01:05:14.050 --> 01:05:19.100 align:middle line:90% 01:05:19.100 --> 01:05:21.140 align:middle line:84% Remember, I said a few minutes ago, 01:05:21.140 --> 01:05:23.500 align:middle line:84% under what condition they were equal. 01:05:23.500 --> 01:05:26.260 align:middle line:84% So that tells you under what condition they're not equal. 01:05:26.260 --> 01:05:26.780 align:middle line:90% Yeah. 01:05:26.780 --> 01:05:27.860 align:middle line:90% AUDIENCE: Imperfect competition. 01:05:27.860 --> 01:05:29.277 align:middle line:84% JON GRUBER: Imperfect competition. 01:05:29.277 --> 01:05:32.200 align:middle line:84% Under perfect competition, opportunity costs, 01:05:32.200 --> 01:05:33.200 align:middle line:90% cash costs are the same. 01:05:33.200 --> 01:05:35.780 align:middle line:84% But imagine the asphalt is produced by a monopoly asphalt 01:05:35.780 --> 01:05:37.300 align:middle line:90% producer. 01:05:37.300 --> 01:05:41.180 align:middle line:84% In that case, the monopolist, the marginal cost, 01:05:41.180 --> 01:05:43.400 align:middle line:84% the monopolist is below the price they're paying. 01:05:43.400 --> 01:05:45.608 align:middle line:84% Remember, for monopolists, price above marginal cost, 01:05:45.608 --> 01:05:47.300 align:middle line:90% we call it the markup. 01:05:47.300 --> 01:05:49.603 align:middle line:84% So the price you pay the monopolist, 01:05:49.603 --> 01:05:51.520 align:middle line:84% minus the marginal cost of the bag of asphalt, 01:05:51.520 --> 01:05:53.620 align:middle line:90% is a transfer to the monopolist. 01:05:53.620 --> 01:05:58.860 align:middle line:84% It's just like the better union job is a transfer to the worker. 01:05:58.860 --> 01:06:01.380 align:middle line:84% The difference between the price of asphalt 01:06:01.380 --> 01:06:05.860 align:middle line:84% and the marginal cost of asphalt is a transfer to the monopolist. 01:06:05.860 --> 01:06:08.180 align:middle line:84% So you would also have an efficiency cost different 01:06:08.180 --> 01:06:09.760 align:middle line:90% from the cash cost in that case. 01:06:09.760 --> 01:06:12.340 align:middle line:90% 01:06:12.340 --> 01:06:12.920 align:middle line:90% Yeah. 01:06:12.920 --> 01:06:13.220 align:middle line:90% AUDIENCE: Sorry. 01:06:13.220 --> 01:06:14.800 align:middle line:84% Can you explain that one more time, 01:06:14.800 --> 01:06:16.732 align:middle line:90% the [INAUDIBLE] monopolists? 01:06:16.732 --> 01:06:19.080 align:middle line:84% JON GRUBER: Well, what is the efficiency 01:06:19.080 --> 01:06:21.118 align:middle line:90% cost of a bag of asphalt? 01:06:21.118 --> 01:06:22.160 align:middle line:90% Here's the key intuition. 01:06:22.160 --> 01:06:23.415 align:middle line:90% I should say it this way. 01:06:23.415 --> 01:06:25.040 align:middle line:84% The efficiency cost is that I'm pulling 01:06:25.040 --> 01:06:29.160 align:middle line:84% that bag of asphalt out of its next best use in the economy. 01:06:29.160 --> 01:06:31.100 align:middle line:84% What is the value of its next best use? 01:06:31.100 --> 01:06:35.600 align:middle line:84% Well, it's the marginal cost of producing a bag of asphalt. What 01:06:35.600 --> 01:06:36.340 align:middle line:90% do we pay for it? 01:06:36.340 --> 01:06:38.435 align:middle line:90% We pay the price. 01:06:38.435 --> 01:06:40.560 align:middle line:84% In a perfectly competitive market, those are equal. 01:06:40.560 --> 01:06:42.020 align:middle line:84% In a perfectly competitive market, they're not. 01:06:42.020 --> 01:06:44.180 align:middle line:84% And the gap is a transfer given to the monopolist. 01:06:44.180 --> 01:06:47.298 align:middle line:90% 01:06:47.298 --> 01:06:48.340 align:middle line:90% Good clarifying question. 01:06:48.340 --> 01:06:49.140 align:middle line:90% Yeah. 01:06:49.140 --> 01:06:50.680 align:middle line:84% AUDIENCE: Is the next best benefit 01:06:50.680 --> 01:06:52.960 align:middle line:84% to society of that bag of asphalt 01:06:52.960 --> 01:06:54.920 align:middle line:84% equal the marginal cost even in the case where 01:06:54.920 --> 01:06:57.120 align:middle line:84% it's a monopolist who-- even if they're not 01:06:57.120 --> 01:07:01.080 align:middle line:84% selling to the government, they're still charging more? 01:07:01.080 --> 01:07:02.580 align:middle line:84% JON GRUBER: That's a great question, 01:07:02.580 --> 01:07:04.705 align:middle line:84% which is, what if the whole industry's monopolized? 01:07:04.705 --> 01:07:08.520 align:middle line:84% Then things get more difficult. You're right. 01:07:08.520 --> 01:07:11.160 align:middle line:84% So I'm assuming there is an efficient market in which it 01:07:11.160 --> 01:07:12.340 align:middle line:90% could otherwise be used. 01:07:12.340 --> 01:07:14.870 align:middle line:90% 01:07:14.870 --> 01:07:15.390 align:middle line:90% Yeah. 01:07:15.390 --> 01:07:19.230 align:middle line:84% AUDIENCE: I have a question about the imperfect competition. 01:07:19.230 --> 01:07:21.850 align:middle line:84% What if, instead of there only being one manufacturer, 01:07:21.850 --> 01:07:24.390 align:middle line:84% there's only one purchaser, like a monopsony? 01:07:24.390 --> 01:07:25.990 align:middle line:84% JON GRUBER: Well, I mean, then you 01:07:25.990 --> 01:07:27.518 align:middle line:84% get into this question asked here, 01:07:27.518 --> 01:07:29.810 align:middle line:84% which is there's only one purchaser for these projects. 01:07:29.810 --> 01:07:32.102 align:middle line:84% Why do they pay him $20 an hour instead of $10 an hour? 01:07:32.102 --> 01:07:34.390 align:middle line:84% You'd think in that case they'd pay below. 01:07:34.390 --> 01:07:37.170 align:middle line:90% But they can't necessarily. 01:07:37.170 --> 01:07:39.910 align:middle line:90% They might not necessarily. 01:07:39.910 --> 01:07:41.950 align:middle line:84% This actually goes back to last lecture, 01:07:41.950 --> 01:07:44.720 align:middle line:84% these examples talk about contracting out. 01:07:44.720 --> 01:07:46.470 align:middle line:84% Part of the reason the public sector often 01:07:46.470 --> 01:07:48.190 align:middle line:84% pays too much for goods is because they 01:07:48.190 --> 01:07:49.925 align:middle line:90% have non-compete contracts. 01:07:49.925 --> 01:07:52.050 align:middle line:84% Now, you think of a non-compete contract for a good 01:07:52.050 --> 01:07:53.650 align:middle line:84% if I'm a monopsony buyer of a missile. 01:07:53.650 --> 01:07:55.070 align:middle line:90% No one else is buying missiles. 01:07:55.070 --> 01:07:56.970 align:middle line:84% I should be able to pay much less. 01:07:56.970 --> 01:07:58.550 align:middle line:84% But in fact, by non-compete, I often 01:07:58.550 --> 01:08:01.202 align:middle line:84% end up paying more because I don't really 01:08:01.202 --> 01:08:02.410 align:middle line:90% know what the right price is. 01:08:02.410 --> 01:08:04.530 align:middle line:84% So when they compete, they reveal the right price to me. 01:08:04.530 --> 01:08:06.070 align:middle line:84% When there's a non-compete, I just 01:08:06.070 --> 01:08:07.487 align:middle line:84% have to go with what they tell me. 01:08:07.487 --> 01:08:10.190 align:middle line:84% And I've got no idea what the right answer is. 01:08:10.190 --> 01:08:12.740 align:middle line:90% All right, other questions? 01:08:12.740 --> 01:08:14.195 align:middle line:90% So that's one-- yeah. 01:08:14.195 --> 01:08:15.820 align:middle line:84% AUDIENCE: Is this non-competes in terms 01:08:15.820 --> 01:08:18.637 align:middle line:84% of, I don't know, McKinsey-- or not McKinsey, sorry, Lockheed? 01:08:18.637 --> 01:08:19.720 align:middle line:90% JON GRUBER: Yeah, exactly. 01:08:19.720 --> 01:08:22.380 align:middle line:84% So basically, I'm buying a missile. 01:08:22.380 --> 01:08:23.979 align:middle line:84% What I can do is say, look, anybody 01:08:23.979 --> 01:08:27.420 align:middle line:84% who produces a missile bid on this missile with this specs. 01:08:27.420 --> 01:08:29.460 align:middle line:84% And unlike with prisons, the specs 01:08:29.460 --> 01:08:31.220 align:middle line:90% can be really pretty precise. 01:08:31.220 --> 01:08:34.380 align:middle line:90% Build a missile to these specs. 01:08:34.380 --> 01:08:36.660 align:middle line:84% I'll get the bids, and I'll pick the lowest one. 01:08:36.660 --> 01:08:38.700 align:middle line:84% That's the way you'd think we'd do it. 01:08:38.700 --> 01:08:42.040 align:middle line:84% That is less than half of all contracts the government does. 01:08:42.040 --> 01:08:44.520 align:middle line:84% And more than half all contracts just say, hey, I like you. 01:08:44.520 --> 01:08:48.082 align:middle line:84% You produce the missile, which has the problem that there's 01:08:48.082 --> 01:08:49.540 align:middle line:84% no competitive pressure and there's 01:08:49.540 --> 01:08:51.060 align:middle line:90% no information revelation. 01:08:51.060 --> 01:08:53.580 align:middle line:84% Competition, remember, is beautiful in that it solves 01:08:53.580 --> 01:08:55.757 align:middle line:84% the information revelation problem, 01:08:55.757 --> 01:08:57.840 align:middle line:84% because competition reveals what marginal cost is. 01:08:57.840 --> 01:08:59.100 align:middle line:84% If you don't have competition, how 01:08:59.100 --> 01:09:00.975 align:middle line:84% do you reveal what the true marginal cost is? 01:09:00.975 --> 01:09:01.500 align:middle line:90% You don't. 01:09:01.500 --> 01:09:03.279 align:middle line:84% And that's the problem with the non-compete contracts. 01:09:03.279 --> 01:09:04.140 align:middle line:90% Why? 01:09:04.140 --> 01:09:04.859 align:middle line:90% Chapter 9. 01:09:04.859 --> 01:09:11.399 align:middle line:90% 01:09:11.399 --> 01:09:13.880 align:middle line:90% That's the first challenge. 01:09:13.880 --> 01:09:17.080 align:middle line:84% So costs, you would think, would be pretty straightforward. 01:09:17.080 --> 01:09:20.560 align:middle line:84% The first challenge is imperfect markets. 01:09:20.560 --> 01:09:25.279 align:middle line:84% The second challenge is discounting, 01:09:25.279 --> 01:09:30.120 align:middle line:84% the fact that not all costs come at the same time. 01:09:30.120 --> 01:09:32.800 align:middle line:84% Now, we haven't covered chapter 4 yet. 01:09:32.800 --> 01:09:34.415 align:middle line:84% But I'm going to assume you remember 01:09:34.415 --> 01:09:36.040 align:middle line:84% the concept of present discounted value 01:09:36.040 --> 01:09:40.680 align:middle line:84% from 14.01, which is that dollars in the future 01:09:40.680 --> 01:09:43.062 align:middle line:84% are less valuable than dollars today. 01:09:43.062 --> 01:09:45.020 align:middle line:84% It's a fundamental concept of discounted value. 01:09:45.020 --> 01:09:47.340 align:middle line:84% So if you want to express the cost of something, 01:09:47.340 --> 01:09:50.920 align:middle line:84% you have to express it in today's dollars, discounting 01:09:50.920 --> 01:09:52.779 align:middle line:84% future dollars because they're worth less. 01:09:52.779 --> 01:09:52.840 align:middle line:90% Why? 01:09:52.840 --> 01:09:54.180 align:middle line:84% They're worth less, because you had them today, 01:09:54.180 --> 01:09:56.800 align:middle line:84% you could invest them and they'd be more tomorrow. 01:09:56.800 --> 01:09:59.700 align:middle line:84% Once again, we'll cover this in chapter 4 to review. 01:09:59.700 --> 01:10:01.920 align:middle line:90% But that was in 14.01. 01:10:01.920 --> 01:10:07.200 align:middle line:84% So basically, the problem is that some of these costs 01:10:07.200 --> 01:10:10.570 align:middle line:84% come in the future, like the maintenance costs. 01:10:10.570 --> 01:10:12.590 align:middle line:90% That's $10 per year. 01:10:12.590 --> 01:10:14.490 align:middle line:84% So we can't-- $10 million a year. 01:10:14.490 --> 01:10:18.290 align:middle line:84% We can't just add that because it's happening in the future, 01:10:18.290 --> 01:10:19.410 align:middle line:90% so it's worth less. 01:10:19.410 --> 01:10:22.810 align:middle line:84% We have to discount it and get a PDV of costs. 01:10:22.810 --> 01:10:25.190 align:middle line:84% The asphalt and the laborers, we hire today. 01:10:25.190 --> 01:10:26.950 align:middle line:90% So there's no PDV issues. 01:10:26.950 --> 01:10:29.610 align:middle line:84% The maintenance, we do over time. 01:10:29.610 --> 01:10:31.870 align:middle line:90% And so we have to discount it. 01:10:31.870 --> 01:10:33.597 align:middle line:84% So then we know how to do discounts. 01:10:33.597 --> 01:10:34.930 align:middle line:90% We know how to PDV calculations. 01:10:34.930 --> 01:10:36.230 align:middle line:90% We've done those. 01:10:36.230 --> 01:10:40.748 align:middle line:84% But the question is, what's the right discount rate to use? 01:10:40.748 --> 01:10:43.290 align:middle line:84% Because now it's a government having to pick a discount rate, 01:10:43.290 --> 01:10:46.110 align:middle line:84% not a private market [INAUDIBLE] discount rate. 01:10:46.110 --> 01:10:50.550 align:middle line:84% Now, you might say what would be a natural starting point 01:10:50.550 --> 01:10:54.510 align:middle line:84% for the right discount rate to use for a government deciding 01:10:54.510 --> 01:10:55.890 align:middle line:90% on whether to do a project. 01:10:55.890 --> 01:10:58.510 align:middle line:90% 01:10:58.510 --> 01:10:59.050 align:middle line:90% Yeah. 01:10:59.050 --> 01:11:02.070 align:middle line:90% AUDIENCE: The inflation rate? 01:11:02.070 --> 01:11:04.923 align:middle line:84% JON GRUBER: Well, no, because the opportunity cost of money 01:11:04.923 --> 01:11:06.090 align:middle line:90% is not just about inflation. 01:11:06.090 --> 01:11:07.840 align:middle line:84% It's about the fact that money could be worth more tomorrow. 01:11:07.840 --> 01:11:09.760 align:middle line:84% Even with zero inflation, it's still a discount rate. 01:11:09.760 --> 01:11:10.300 align:middle line:90% Yeah. 01:11:10.300 --> 01:11:11.300 align:middle line:90% AUDIENCE: Cost of borrowing. 01:11:11.300 --> 01:11:13.260 align:middle line:84% JON GRUBER: The private sector discount rate-- 01:11:13.260 --> 01:11:15.260 align:middle line:84% a natural point to say, look, the private sector 01:11:15.260 --> 01:11:16.320 align:middle line:90% is a cost of capital. 01:11:16.320 --> 01:11:17.120 align:middle line:90% It's what it costs. 01:11:17.120 --> 01:11:17.940 align:middle line:90% Now, they don't just borrow. 01:11:17.940 --> 01:11:19.280 align:middle line:90% They issue debt and equity. 01:11:19.280 --> 01:11:21.540 align:middle line:84% We have an aggregate cost of capital. 01:11:21.540 --> 01:11:23.380 align:middle line:90% It's about 7%. 01:11:23.380 --> 01:11:25.840 align:middle line:90% It's an historical estimate. 01:11:25.840 --> 01:11:27.860 align:middle line:84% So for a typical firm needing to raise capital 01:11:27.860 --> 01:11:30.340 align:middle line:84% through equity or debt or other means, in aggregate, 01:11:30.340 --> 01:11:32.540 align:middle line:90% it's about a 7% cost of capital. 01:11:32.540 --> 01:11:36.900 align:middle line:84% You have to pay about a 7% rate of return to get your money. 01:11:36.900 --> 01:11:39.620 align:middle line:84% So a natural starting point is to say, look-- 01:11:39.620 --> 01:11:41.580 align:middle line:84% and this is why the starting point makes sense. 01:11:41.580 --> 01:11:45.340 align:middle line:84% If I take that money to build this highway, 01:11:45.340 --> 01:11:47.280 align:middle line:84% what's the alternative use of the money? 01:11:47.280 --> 01:11:49.535 align:middle line:84% The private sector could have had it. 01:11:49.535 --> 01:11:51.160 align:middle line:84% So if I'm spending money in the future, 01:11:51.160 --> 01:11:54.700 align:middle line:84% I should use the private sector discount rate. 01:11:54.700 --> 01:11:59.740 align:middle line:84% But there's one big difference, which is subtle but important. 01:11:59.740 --> 01:12:04.920 align:middle line:84% 7% is the pre-tax historical rate of return on capital. 01:12:04.920 --> 01:12:08.920 align:middle line:84% But if you save in the bank, you get taxed on that. 01:12:08.920 --> 01:12:13.160 align:middle line:84% Post-tax, it's more like 5% or whatever it is. 01:12:13.160 --> 01:12:17.120 align:middle line:84% So the rate of return on capital used by the private sector 01:12:17.120 --> 01:12:19.880 align:middle line:90% is the post-tax rate of return. 01:12:19.880 --> 01:12:23.240 align:middle line:84% Should that be the rate used by the government? 01:12:23.240 --> 01:12:24.152 align:middle line:90% Why or why not? 01:12:24.152 --> 01:12:26.360 align:middle line:84% Should the government follow the private sector using 01:12:26.360 --> 01:12:28.180 align:middle line:90% the post-tax rate of return? 01:12:28.180 --> 01:12:28.680 align:middle line:90% Yeah. 01:12:28.680 --> 01:12:31.020 align:middle line:84% AUDIENCE: No, because they're not taxing themselves. 01:12:31.020 --> 01:12:32.720 align:middle line:90% JON GRUBER: They keep the money. 01:12:32.720 --> 01:12:34.160 align:middle line:84% They get the tax revenue, so they 01:12:34.160 --> 01:12:38.080 align:middle line:84% should use-- the right answer is that the federal government 01:12:38.080 --> 01:12:40.600 align:middle line:84% should discount its future costs and future 01:12:40.600 --> 01:12:46.305 align:middle line:84% benefits by the pre-tax private cost of capital. 01:12:46.305 --> 01:12:48.680 align:middle line:84% Because the alternative is, private sector has the money, 01:12:48.680 --> 01:12:52.160 align:middle line:84% and that's their discount rate that they use. 01:12:52.160 --> 01:12:55.560 align:middle line:84% Now, the government uses a variety of different discount 01:12:55.560 --> 01:12:59.260 align:middle line:84% rates, but they're centered around 7%. 01:12:59.260 --> 01:13:05.910 align:middle line:84% So let's use 7% Let's now go to table 8.2 01:13:05.910 --> 01:13:10.950 align:middle line:90% and let's look at the costs. 01:13:10.950 --> 01:13:13.610 align:middle line:84% We have a million bags of asphalt. 01:13:13.610 --> 01:13:15.650 align:middle line:84% Let's assume that the price of asphalt is $100. 01:13:15.650 --> 01:13:17.870 align:middle line:84% Let's assume that's a perfectly competitive market. 01:13:17.870 --> 01:13:19.850 align:middle line:84% So the opportunity cost is the cash cost. 01:13:19.850 --> 01:13:20.770 align:middle line:90% It's $100 a bag. 01:13:20.770 --> 01:13:22.830 align:middle line:90% That's $100 million. 01:13:22.830 --> 01:13:24.530 align:middle line:84% We have a million of hours of labor. 01:13:24.530 --> 01:13:26.988 align:middle line:84% But we're going to use the opportunity cost of labor, which 01:13:26.988 --> 01:13:29.390 align:middle line:90% is the market wage of $10. 01:13:29.390 --> 01:13:31.950 align:middle line:90% So that's $10 million. 01:13:31.950 --> 01:13:35.190 align:middle line:84% We have maintenance of $10 million a year forever, 01:13:35.190 --> 01:13:37.450 align:middle line:90% but there's a 7% discount rate. 01:13:37.450 --> 01:13:39.990 align:middle line:84% You all remember the formula that, if you 01:13:39.990 --> 01:13:42.590 align:middle line:84% have an infinite sum, the value of something 01:13:42.590 --> 01:13:46.990 align:middle line:84% is the amount you're paying out, F, over the interest rate. 01:13:46.990 --> 01:13:55.870 align:middle line:84% So $10 million over 0.07 is $143 million. 01:13:55.870 --> 01:13:57.250 align:middle line:90% That is the three costs. 01:13:57.250 --> 01:14:01.900 align:middle line:84% So the total cost of this effort in today's dollars 01:14:01.900 --> 01:14:04.500 align:middle line:90% is $253 million. 01:14:04.500 --> 01:14:05.640 align:middle line:90% Questions about that? 01:14:05.640 --> 01:14:08.740 align:middle line:90% 01:14:08.740 --> 01:14:10.820 align:middle line:84% Now we're going to go to the harder 01:14:10.820 --> 01:14:13.757 align:middle line:90% part, which is the benefits. 01:14:13.757 --> 01:14:15.340 align:middle line:84% Let me just introduce this, then we'll 01:14:15.340 --> 01:14:17.340 align:middle line:90% stop and get into it next time. 01:14:17.340 --> 01:14:21.980 align:middle line:84% The benefits are harder because they value things like time 01:14:21.980 --> 01:14:23.420 align:middle line:90% and lives. 01:14:23.420 --> 01:14:25.860 align:middle line:84% These are things without market prices. 01:14:25.860 --> 01:14:28.020 align:middle line:84% We can talk about the flaws in the market 01:14:28.020 --> 01:14:29.800 align:middle line:90% prices for asphalt and labor. 01:14:29.800 --> 01:14:33.060 align:middle line:84% But at least there are market prices. 01:14:33.060 --> 01:14:36.240 align:middle line:84% Time and lives do not have market prices. 01:14:36.240 --> 01:14:39.900 align:middle line:84% Now time, you might say, has a market price, which is the wage. 01:14:39.900 --> 01:14:43.000 align:middle line:84% And that would be a natural starting point valuing time. 01:14:43.000 --> 01:14:45.500 align:middle line:84% It would be to say, look, we should value 01:14:45.500 --> 01:14:49.820 align:middle line:90% the time you save by your wage. 01:14:49.820 --> 01:14:54.220 align:middle line:84% And that would be what we call a market-based solution. 01:14:54.220 --> 01:14:56.820 align:middle line:84% A market-based solution to valuing time 01:14:56.820 --> 01:15:00.320 align:middle line:90% would be to use your wage. 01:15:00.320 --> 01:15:04.280 align:middle line:84% Now suppose I said to you, well, you 01:15:04.280 --> 01:15:07.200 align:middle line:84% can't use a wage to value time because people 01:15:07.200 --> 01:15:10.920 align:middle line:90% don't work all the time. 01:15:10.920 --> 01:15:15.880 align:middle line:84% And you responded to me, in fact, the wage times 24 hours 01:15:15.880 --> 01:15:18.540 align:middle line:84% is, in a perfectly competitive labor market, 01:15:18.540 --> 01:15:21.980 align:middle line:84% the proper way to value time, even if as with most people 01:15:21.980 --> 01:15:25.120 align:middle line:84% they only work about a quarter of the time of their hours. 01:15:25.120 --> 01:15:27.000 align:middle line:90% Why is that? 01:15:27.000 --> 01:15:28.560 align:middle line:90% Yeah. 01:15:28.560 --> 01:15:31.120 align:middle line:84% AUDIENCE: Because the not working is the value of leisure. 01:15:31.120 --> 01:15:31.870 align:middle line:90% JON GRUBER: Right. 01:15:31.870 --> 01:15:34.000 align:middle line:84% In equilibrium, people set the wage 01:15:34.000 --> 01:15:38.260 align:middle line:84% equal to the marginal value of leisure. 01:15:38.260 --> 01:15:39.580 align:middle line:90% We solved that in 14.01. 01:15:39.580 --> 01:15:41.120 align:middle line:90% And think about the intuition. 01:15:41.120 --> 01:15:44.280 align:middle line:84% In a perfectly competitive labor market, 01:15:44.280 --> 01:15:48.260 align:middle line:84% if working is worth less to you than not working, 01:15:48.260 --> 01:15:49.800 align:middle line:90% what would you do? 01:15:49.800 --> 01:15:51.160 align:middle line:90% Work less. 01:15:51.160 --> 01:15:53.260 align:middle line:84% If work was worth more to you than not working, 01:15:53.260 --> 01:15:54.160 align:middle line:90% what would you do? 01:15:54.160 --> 01:15:55.960 align:middle line:90% You'd work more. 01:15:55.960 --> 01:15:59.110 align:middle line:84% So you will work until the point where the wage equals 01:15:59.110 --> 01:16:01.510 align:middle line:90% a marginal value of leisure. 01:16:01.510 --> 01:16:04.510 align:middle line:84% So it doesn't matter if the time you're saving by driving faster 01:16:04.510 --> 01:16:07.270 align:middle line:84% comes from your kid's soccer game or work. 01:16:07.270 --> 01:16:12.030 align:middle line:84% Either way, you value it at your wage. 01:16:12.030 --> 01:16:13.390 align:middle line:90% The problem of-- questions? 01:16:13.390 --> 01:16:15.290 align:middle line:90% People understand that? 01:16:15.290 --> 01:16:16.790 align:middle line:84% The problem, of course with that is, 01:16:16.790 --> 01:16:18.670 align:middle line:84% we don't have a perfect labor market. 01:16:18.670 --> 01:16:20.950 align:middle line:84% And as a result, the wage can actually 01:16:20.950 --> 01:16:23.510 align:middle line:84% be very different than the marginal value of leisure. 01:16:23.510 --> 01:16:29.310 align:middle line:84% So what's an example of why the wage could be less 01:16:29.310 --> 01:16:31.990 align:middle line:84% than the marginal value of leisure? 01:16:31.990 --> 01:16:33.365 align:middle line:84% What would that mean for the wage 01:16:33.365 --> 01:16:35.032 align:middle line:84% to be less than marginal value solution? 01:16:35.032 --> 01:16:35.950 align:middle line:90% Why could that arise? 01:16:35.950 --> 01:16:37.598 align:middle line:84% Let's get some other folks involved. 01:16:37.598 --> 01:16:39.390 align:middle line:84% What would it mean for your wage to be less 01:16:39.390 --> 01:16:40.970 align:middle line:84% than your marginal value of leisure? 01:16:40.970 --> 01:16:44.670 align:middle line:90% And why might that arise? 01:16:44.670 --> 01:16:46.450 align:middle line:90% Take a stab, I don't bite. 01:16:46.450 --> 01:16:46.950 align:middle line:90% Yeah. 01:16:46.950 --> 01:16:49.258 align:middle line:84% AUDIENCE: You could have passive income. 01:16:49.258 --> 01:16:50.750 align:middle line:90% JON GRUBER: Yeah. 01:16:50.750 --> 01:16:53.010 align:middle line:84% Well, a passive income, it's marginal. 01:16:53.010 --> 01:16:55.390 align:middle line:84% So passive income is a fixed other thing. 01:16:55.390 --> 01:16:58.460 align:middle line:90% But let's extend that. 01:16:58.460 --> 01:17:00.320 align:middle line:84% Let's not think of passive income. 01:17:00.320 --> 01:17:03.800 align:middle line:84% What else comes with jobs beside wages? 01:17:03.800 --> 01:17:04.740 align:middle line:90% AUDIENCE: Benefits. 01:17:04.740 --> 01:17:06.060 align:middle line:90% JON GRUBER: Benefits. 01:17:06.060 --> 01:17:08.940 align:middle line:84% And basically, some of those benefits 01:17:08.940 --> 01:17:11.740 align:middle line:84% would be linked to how hard you work. 01:17:11.740 --> 01:17:15.580 align:middle line:84% In fact, the marginal value of work is not the wage. 01:17:15.580 --> 01:17:17.900 align:middle line:90% It's your total compensation. 01:17:17.900 --> 01:17:20.980 align:middle line:84% So you really should be setting compensation 01:17:20.980 --> 01:17:24.940 align:middle line:84% equal to the marginal value of leisure, not wage. 01:17:24.940 --> 01:17:28.580 align:middle line:84% So you should be choosing a higher marginal value 01:17:28.580 --> 01:17:31.660 align:middle line:90% of leisure. 01:17:31.660 --> 01:17:33.220 align:middle line:84% Actually, it's a backward example. 01:17:33.220 --> 01:17:36.060 align:middle line:84% That would say you'd want to work less. 01:17:36.060 --> 01:17:39.300 align:middle line:84% This is an opposite example of what I suggested. 01:17:39.300 --> 01:17:40.900 align:middle line:90% So we'll do the flip next. 01:17:40.900 --> 01:17:42.120 align:middle line:90% You'd want to work less. 01:17:42.120 --> 01:17:42.640 align:middle line:90% Why? 01:17:42.640 --> 01:17:46.680 align:middle line:84% Because a higher marginal value of leisure means less leisure. 01:17:46.680 --> 01:17:49.780 align:middle line:84% Remember, margin's the opposite way we think. 01:17:49.780 --> 01:17:52.020 align:middle line:84% Higher marginal utility means less consumption. 01:17:52.020 --> 01:17:54.460 align:middle line:84% Higher marginal value of leisure means less leisure. 01:17:54.460 --> 01:17:57.880 align:middle line:84% So if you consider benefits, that's 01:17:57.880 --> 01:17:59.600 align:middle line:84% an example where actually, the wage 01:17:59.600 --> 01:18:01.980 align:middle line:84% is going to understate the value of leisure. 01:18:01.980 --> 01:18:04.000 align:middle line:84% What would be an example the wage 01:18:04.000 --> 01:18:06.280 align:middle line:84% might overstate the value of leisure. 01:18:06.280 --> 01:18:06.820 align:middle line:90% Yeah. 01:18:06.820 --> 01:18:08.660 align:middle line:84% AUDIENCE: You might want to work more. 01:18:08.660 --> 01:18:09.460 align:middle line:84% JON GRUBER: You might want to work more. 01:18:09.460 --> 01:18:10.340 align:middle line:90% But why couldn't you? 01:18:10.340 --> 01:18:12.600 align:middle line:84% AUDIENCE: Not enough employment opportunities. 01:18:12.600 --> 01:18:16.180 align:middle line:84% JON GRUBER: Or there's overtime loss. 01:18:16.180 --> 01:18:20.400 align:middle line:84% You set W equals MBL, or MBL at 43.5 hours a week, 01:18:20.400 --> 01:18:22.950 align:middle line:84% and your employer just won't let you have more than 40. 01:18:22.950 --> 01:18:24.700 align:middle line:84% Well, then you're not working hard enough. 01:18:24.700 --> 01:18:27.140 align:middle line:84% Your wage is then above your marginal value of leisure. 01:18:27.140 --> 01:18:30.440 align:middle line:84% But you can't get to the point where they're equal. 01:18:30.440 --> 01:18:34.520 align:middle line:84% So for reasons like these, wage is an imperfect measure 01:18:34.520 --> 01:18:37.480 align:middle line:84% of the marginal value of leisure. 01:18:37.480 --> 01:18:38.620 align:middle line:90% Let me stop there. 01:18:38.620 --> 01:18:39.560 align:middle line:90% We'll come back. 01:18:39.560 --> 01:18:42.240 align:middle line:84% Paul, we should start next time with table 8-2. 01:18:42.240 --> 01:18:46.740 align:middle line:84% And it will do 8-2, 8-3, 8-4, and then the table in Chapter 9. 01:18:46.740 --> 01:18:50.430 align:middle line:84% So we'll finish up chapter 8 and get into chapter 9. 01:18:50.430 --> 01:18:55.000 align:middle line:90%