WEBVTT 00:00:00.000 --> 00:00:01.967 align:middle line:90% [SQUEAKING] 00:00:01.967 --> 00:00:03.923 align:middle line:90% [RUSTLING] 00:00:03.923 --> 00:00:07.346 align:middle line:90% [CLICKING] 00:00:07.346 --> 00:00:09.765 align:middle line:90% 00:00:09.765 --> 00:00:11.640 align:middle line:84% JON GRUBER: Today, we continue our discussion 00:00:11.640 --> 00:00:14.840 align:middle line:84% of social insurance by turning to health care. 00:00:14.840 --> 00:00:17.280 align:middle line:84% Now, health care is the largest single expenditure 00:00:17.280 --> 00:00:18.708 align:middle line:90% of the US federal government. 00:00:18.708 --> 00:00:21.000 align:middle line:84% It's the largest single expenditure of most governments 00:00:21.000 --> 00:00:22.920 align:middle line:90% around the world. 00:00:22.920 --> 00:00:27.040 align:middle line:90% And it really is an area where-- 00:00:27.040 --> 00:00:30.540 align:middle line:84% when we talked about our social insurance programs, 00:00:30.540 --> 00:00:33.520 align:middle line:84% I talked about how you need to know some institutional details 00:00:33.520 --> 00:00:34.860 align:middle line:90% to understand the area. 00:00:34.860 --> 00:00:37.000 align:middle line:90% And I gave you 15 or 20 minutes. 00:00:37.000 --> 00:00:41.320 align:middle line:84% This is you need to know an entire sixth of the economy 00:00:41.320 --> 00:00:44.890 align:middle line:84% to really understand it, which is a lot tougher. 00:00:44.890 --> 00:00:46.640 align:middle line:84% So we're going to just spend a lot of time 00:00:46.640 --> 00:00:47.700 align:middle line:90% in the next few lectures. 00:00:47.700 --> 00:00:49.760 align:middle line:84% We'll spend three lectures on this. 00:00:49.760 --> 00:00:53.880 align:middle line:84% We'll spend a lot of time on the background on health economics, 00:00:53.880 --> 00:00:58.457 align:middle line:84% even to the extent it isn't directly about the public policy 00:00:58.457 --> 00:01:00.290 align:middle line:84% because you really need to understand health 00:01:00.290 --> 00:01:02.370 align:middle line:84% economics to then think appropriately 00:01:02.370 --> 00:01:03.950 align:middle line:90% about the public policy issues. 00:01:03.950 --> 00:01:05.970 align:middle line:84% So what we do in chapter 15 is really 00:01:05.970 --> 00:01:09.010 align:middle line:84% cover the basics of health economics. 00:01:09.010 --> 00:01:12.090 align:middle line:84% It's a semester of health economics 00:01:12.090 --> 00:01:14.250 align:middle line:84% in one chapter covering the things you 00:01:14.250 --> 00:01:18.250 align:middle line:84% need to know to be an educated consumer of health care 00:01:18.250 --> 00:01:20.330 align:middle line:84% policy and producer of health care policy. 00:01:20.330 --> 00:01:23.610 align:middle line:84% Chapter 16 is then the applications 00:01:23.610 --> 00:01:26.650 align:middle line:84% to the biggest government expenditure policies, Medicaid 00:01:26.650 --> 00:01:27.808 align:middle line:90% and Medicare. 00:01:27.808 --> 00:01:29.350 align:middle line:84% So that's how we're going to lay out. 00:01:29.350 --> 00:01:30.790 align:middle line:84% And we'll cover this over three lectures. 00:01:30.790 --> 00:01:32.130 align:middle line:90% There's a lot to go through. 00:01:32.130 --> 00:01:34.880 align:middle line:84% But the introduction, chapter 15 is my favorite introduction 00:01:34.880 --> 00:01:35.630 align:middle line:90% in the whole book. 00:01:35.630 --> 00:01:37.297 align:middle line:84% So I'm just going to read it rather than 00:01:37.297 --> 00:01:38.730 align:middle line:90% trying to summarize it. 00:01:38.730 --> 00:01:42.610 align:middle line:84% On March 30th, 1981, only two months after entering office, 00:01:42.610 --> 00:01:44.908 align:middle line:84% President Ronald Reagan was shot by John Hinckley, 00:01:44.908 --> 00:01:47.450 align:middle line:84% a deranged fan of Jodie Foster, who thought the actress would 00:01:47.450 --> 00:01:50.290 align:middle line:84% pay more attention to him if he killed the president. 00:01:50.290 --> 00:01:52.970 align:middle line:84% Reagan was rushed to George Washington Hospital, where 00:01:52.970 --> 00:01:54.570 align:middle line:84% he underwent three hours of surgery 00:01:54.570 --> 00:01:56.890 align:middle line:84% to remove a bullet that entered his lung. 00:01:56.890 --> 00:01:58.940 align:middle line:84% Though the injury was potentially fatal, 00:01:58.940 --> 00:02:00.620 align:middle line:84% Reagan made a full recovery and went on 00:02:00.620 --> 00:02:02.580 align:middle line:84% to serve a full eight years in office. 00:02:02.580 --> 00:02:06.140 align:middle line:84% 100 years earlier, President James Garfield was not so lucky. 00:02:06.140 --> 00:02:09.280 align:middle line:84% He, too, was shot two months after entering office on July 2, 00:02:09.280 --> 00:02:12.820 align:middle line:84% 1881 by Charles Guiteau, who thought God had ordered 00:02:12.820 --> 00:02:14.420 align:middle line:90% him to kill the president. 00:02:14.420 --> 00:02:17.280 align:middle line:84% Garfield was brought back to the White House, where for 80 days, 00:02:17.280 --> 00:02:19.020 align:middle line:84% a dozen doctors attempted to find 00:02:19.020 --> 00:02:21.200 align:middle line:84% one bullet that, unbeknownst to them, 00:02:21.200 --> 00:02:23.460 align:middle line:84% had lodged itself near Garfield's spine. 00:02:23.460 --> 00:02:25.180 align:middle line:84% The doctors probed the president's wound 00:02:25.180 --> 00:02:27.820 align:middle line:84% with their unsterilized fingers and metal rods, 00:02:27.820 --> 00:02:29.600 align:middle line:84% succeeding only in widening the wound, 00:02:29.600 --> 00:02:31.940 align:middle line:84% infecting it, and puncturing his liver. 00:02:31.940 --> 00:02:33.900 align:middle line:84% Alexander Graham Bell ran a metal detector 00:02:33.900 --> 00:02:35.780 align:middle line:84% over Garfield's body and soon announced 00:02:35.780 --> 00:02:37.155 align:middle line:90% that he found the bullet. 00:02:37.155 --> 00:02:38.780 align:middle line:84% Surgeons went to work, but still failed 00:02:38.780 --> 00:02:41.020 align:middle line:84% to locate it, not realizing that bell had mistakenly 00:02:41.020 --> 00:02:43.940 align:middle line:84% detected one of the bedsprings underneath the president. 00:02:43.940 --> 00:02:46.220 align:middle line:84% The infection resulting from his poor medical care 00:02:46.220 --> 00:02:48.153 align:middle line:84% soon caused Garfield to have a heart attack. 00:02:48.153 --> 00:02:49.820 align:middle line:84% And when he eventually died, his coroner 00:02:49.820 --> 00:02:51.903 align:middle line:84% declared that Garfield would have survived if only 00:02:51.903 --> 00:02:53.620 align:middle line:90% his doctors left him alone. 00:02:53.620 --> 00:02:55.040 align:middle line:90% So that's 100 years. 00:02:55.040 --> 00:02:56.950 align:middle line:84% That's a big difference over 100 years. 00:02:56.950 --> 00:02:58.950 align:middle line:84% And why I love that story, it summarizes 00:02:58.950 --> 00:03:02.570 align:middle line:84% the unbelievable change in health care in the US 00:03:02.570 --> 00:03:04.570 align:middle line:90% over a hundred-year period. 00:03:04.570 --> 00:03:07.190 align:middle line:84% And it's continued, of course, since 1980 over the next 45 00:03:07.190 --> 00:03:08.690 align:middle line:90% years to continue to improve. 00:03:08.690 --> 00:03:12.370 align:middle line:84% But the bottom line is, at the turn of the 20th century, 00:03:12.370 --> 00:03:14.990 align:middle line:84% health services were basically a hospital someplace you went 00:03:14.990 --> 00:03:15.693 align:middle line:90% to die. 00:03:15.693 --> 00:03:18.110 align:middle line:84% Health care was something where essentially it was a 50/50 00:03:18.110 --> 00:03:20.110 align:middle line:84% proposition, whether it help or not. 00:03:20.110 --> 00:03:22.590 align:middle line:84% By 1980, we had someone shot in their lung 00:03:22.590 --> 00:03:26.830 align:middle line:84% who was surviving with the best health in the country. 00:03:26.830 --> 00:03:30.183 align:middle line:84% Basically, health care has changed enormously. 00:03:30.183 --> 00:03:31.850 align:middle line:84% And there's a couple ways to look at it. 00:03:31.850 --> 00:03:33.767 align:middle line:84% One way to look at is how much we spend on it. 00:03:33.767 --> 00:03:36.970 align:middle line:84% So in 1950, when we started really tracking this stuff, 00:03:36.970 --> 00:03:39.910 align:middle line:84% health care was about 5% of the US economy. 00:03:39.910 --> 00:03:42.990 align:middle line:84% So in 1950, it was 5% of this economy. 00:03:42.990 --> 00:03:47.030 align:middle line:84% That was less than we spent on cars, fuel, or clothing. 00:03:47.030 --> 00:03:49.870 align:middle line:84% Today, health care is 18% of the US 00:03:49.870 --> 00:03:53.830 align:middle line:84% economy, which is more than we spend on housing or food. 00:03:53.830 --> 00:03:56.170 align:middle line:90% And it's not slowing down. 00:03:56.170 --> 00:03:57.930 align:middle line:84% It's been growing historically at a rate 00:03:57.930 --> 00:04:00.010 align:middle line:84% well above the growth rate of GDP, 00:04:00.010 --> 00:04:03.710 align:middle line:84% so as GDP share continues to rise. 00:04:03.710 --> 00:04:06.990 align:middle line:84% Now we might say, gee, that's terrible. 00:04:06.990 --> 00:04:08.030 align:middle line:90% It's going up so fast. 00:04:08.030 --> 00:04:10.810 align:middle line:84% But it's not necessarily bad because health care has 00:04:10.810 --> 00:04:12.190 align:middle line:90% gotten really good in America. 00:04:12.190 --> 00:04:14.250 align:middle line:90% Health care in 1950 sucked. 00:04:14.250 --> 00:04:18.190 align:middle line:84% If you had a baby that was born low birth weight, 00:04:18.190 --> 00:04:21.730 align:middle line:84% it was four times as likely to die as one born today. 00:04:21.730 --> 00:04:23.450 align:middle line:84% If you had a heart attack, you were 00:04:23.450 --> 00:04:25.370 align:middle line:84% four times as likely to die as you are today 00:04:25.370 --> 00:04:27.450 align:middle line:84% or more relevant for your age group. 00:04:27.450 --> 00:04:29.308 align:middle line:90% If you hurt your knee-- 00:04:29.308 --> 00:04:30.350 align:middle line:90% Enoch just hurt his knee. 00:04:30.350 --> 00:04:32.790 align:middle line:84% If you hurt your knee, more serious than you, 00:04:32.790 --> 00:04:35.210 align:middle line:84% if you tore your ACL in 1950, you 00:04:35.210 --> 00:04:36.728 align:middle line:84% were in the hospital for six weeks, 00:04:36.728 --> 00:04:38.270 align:middle line:84% or you're in the hospital for a week, 00:04:38.270 --> 00:04:39.370 align:middle line:84% you were on crutches for six weeks, 00:04:39.370 --> 00:04:41.330 align:middle line:84% and you limp for the rest of your life. 00:04:41.330 --> 00:04:43.570 align:middle line:84% Today, the outpatient surgery, you're 00:04:43.570 --> 00:04:46.050 align:middle line:90% back on the slopes a week later. 00:04:46.050 --> 00:04:50.990 align:middle line:84% Health care is just much, much better than it was in 1950. 00:04:50.990 --> 00:04:54.420 align:middle line:84% It's also much, much more expensive. 00:04:54.420 --> 00:05:01.420 align:middle line:84% So at the same time, we have a country with enormous health 00:05:01.420 --> 00:05:02.980 align:middle line:90% care disparities. 00:05:02.980 --> 00:05:05.660 align:middle line:84% So a common fact you'll hear is that, well, US health care 00:05:05.660 --> 00:05:08.120 align:middle line:84% is really expensive, but our outcomes are terrible. 00:05:08.120 --> 00:05:10.580 align:middle line:84% We're like 19th in the world infant mortality. 00:05:10.580 --> 00:05:13.660 align:middle line:84% That is an incredibly misleading statistic. 00:05:13.660 --> 00:05:16.380 align:middle line:84% A white baby born in America today 00:05:16.380 --> 00:05:19.260 align:middle line:84% has as good outcomes as anywhere else in the world. 00:05:19.260 --> 00:05:21.180 align:middle line:84% A Black baby born today has worse outcomes 00:05:21.180 --> 00:05:23.260 align:middle line:90% than one born in Barbados. 00:05:23.260 --> 00:05:26.540 align:middle line:84% So the problem in the US is not that it's 00:05:26.540 --> 00:05:28.040 align:middle line:90% bad health care for everyone. 00:05:28.040 --> 00:05:29.910 align:middle line:84% The problem is the disparities that 00:05:29.910 --> 00:05:31.660 align:middle line:84% lead them to be truly horrific health care 00:05:31.660 --> 00:05:34.420 align:middle line:84% outcomes for some groups and pretty good health 00:05:34.420 --> 00:05:37.340 align:middle line:90% outcomes for others. 00:05:37.340 --> 00:05:41.460 align:middle line:84% Now, basically, we want to talk about is how do we-- 00:05:41.460 --> 00:05:43.780 align:middle line:84% and moreover, the last fact and the most important fact 00:05:43.780 --> 00:05:45.860 align:middle line:84% for our course is the US health care 00:05:45.860 --> 00:05:49.600 align:middle line:84% system is almost a perfectly split private-public system. 00:05:49.600 --> 00:05:51.630 align:middle line:84% Almost exactly 50% of the spending 00:05:51.630 --> 00:05:55.090 align:middle line:84% is done by the public sector and 50% by the private sector, 00:05:55.090 --> 00:05:57.710 align:middle line:84% within a few percentage points on either side. 00:05:57.710 --> 00:06:00.710 align:middle line:84% So we have a situation where we're starting 50/50. 00:06:00.710 --> 00:06:03.790 align:middle line:84% And the question in front of us is should the government 00:06:03.790 --> 00:06:06.510 align:middle line:90% doing more, less or different? 00:06:06.510 --> 00:06:08.430 align:middle line:84% And that's what we really want to think about 00:06:08.430 --> 00:06:10.470 align:middle line:90% as we talk about health care. 00:06:10.470 --> 00:06:12.213 align:middle line:84% To do that, we need to really understand 00:06:12.213 --> 00:06:13.130 align:middle line:90% some health economics. 00:06:13.130 --> 00:06:14.922 align:middle line:84% We need to understand how health care works 00:06:14.922 --> 00:06:17.530 align:middle line:84% before we can talk intelligently about the government's role. 00:06:17.530 --> 00:06:21.710 align:middle line:84% So like I said, we'll start with some of health economics basics, 00:06:21.710 --> 00:06:24.990 align:middle line:84% and then we'll turn to talking about what the government does. 00:06:24.990 --> 00:06:27.170 align:middle line:84% So let's start with the US health care system. 00:06:27.170 --> 00:06:32.270 align:middle line:84% Figure 15-1 shows health spending as a percent of GDP 00:06:32.270 --> 00:06:36.510 align:middle line:84% as of 2019 across developed nations. 00:06:36.510 --> 00:06:40.310 align:middle line:84% As you can see, the US really stands out as an outlier. 00:06:40.310 --> 00:06:43.430 align:middle line:84% There's a pretty smooth rise and then 00:06:43.430 --> 00:06:46.910 align:middle line:84% this jump from the second most expensive country to the most 00:06:46.910 --> 00:06:49.590 align:middle line:84% expensive country, which is the US. 00:06:49.590 --> 00:06:50.860 align:middle line:90% Where does our money go? 00:06:50.860 --> 00:06:53.897 align:middle line:84% Well, then you can go to figure 15-2. 00:06:53.897 --> 00:06:56.480 align:middle line:84% About a third of the money goes to hospitals and about a third 00:06:56.480 --> 00:06:57.260 align:middle line:90% goes to doctors. 00:06:57.260 --> 00:06:59.880 align:middle line:84% A little more to hospitals than doctors, but roughly speaking, 00:06:59.880 --> 00:07:03.420 align:middle line:84% it's a third hospitals, a third doctors, a third everything 00:07:03.420 --> 00:07:03.920 align:middle line:90% else. 00:07:03.920 --> 00:07:07.400 align:middle line:84% Of the everything else, prescription drugs is about 12%. 00:07:07.400 --> 00:07:11.060 align:middle line:84% That is a rapidly rising share over time, 00:07:11.060 --> 00:07:15.280 align:middle line:84% but still a minority of health care spending. 00:07:15.280 --> 00:07:17.520 align:middle line:90% So that's where our dollars go. 00:07:17.520 --> 00:07:21.160 align:middle line:84% The question is, why do we spend so many dollars? 00:07:21.160 --> 00:07:23.400 align:middle line:84% Why do we have health care spending 00:07:23.400 --> 00:07:25.380 align:middle line:84% that's so much more than the rest of the world? 00:07:25.380 --> 00:07:28.255 align:middle line:84% And to be fair, even for people like myself 00:07:28.255 --> 00:07:29.880 align:middle line:84% who get great health care, our outcomes 00:07:29.880 --> 00:07:31.838 align:middle line:84% still aren't better than the rest of the world. 00:07:31.838 --> 00:07:32.640 align:middle line:90% They're comparable. 00:07:32.640 --> 00:07:35.140 align:middle line:84% They're terrible for people who don't have good health care. 00:07:35.140 --> 00:07:36.380 align:middle line:84% But even for those who have good health care, 00:07:36.380 --> 00:07:38.240 align:middle line:84% outcomes are comparable, not better. 00:07:38.240 --> 00:07:40.320 align:middle line:90% So what's the inefficiency? 00:07:40.320 --> 00:07:41.960 align:middle line:90% We are spending more. 00:07:41.960 --> 00:07:44.060 align:middle line:84% We're getting the same outcomes at best. 00:07:44.060 --> 00:07:45.800 align:middle line:90% What's going on? 00:07:45.800 --> 00:07:51.570 align:middle line:84% Well, basically, essentially, we can break this down 00:07:51.570 --> 00:07:54.375 align:middle line:90% into several different things. 00:07:54.375 --> 00:07:55.250 align:middle line:90% One way to start is-- 00:07:55.250 --> 00:07:56.750 align:middle line:90% I want to look at figure 15-3. 00:07:56.750 --> 00:07:59.450 align:middle line:84% This gives us an international comparison. 00:07:59.450 --> 00:08:02.962 align:middle line:84% It just picks five countries, five developed countries. 00:08:02.962 --> 00:08:05.170 align:middle line:84% What we see is the US has the highest health spending 00:08:05.170 --> 00:08:07.018 align:middle line:90% per person by a large amount. 00:08:07.018 --> 00:08:09.310 align:middle line:84% We also have the worst death rate, although, as I said, 00:08:09.310 --> 00:08:11.610 align:middle line:84% that's really driven by disparities. 00:08:11.610 --> 00:08:13.710 align:middle line:84% We have the highest preventable death rate. 00:08:13.710 --> 00:08:15.210 align:middle line:84% We have the largest number of people 00:08:15.210 --> 00:08:18.290 align:middle line:84% that went without health care because of cost of last year 00:08:18.290 --> 00:08:20.190 align:middle line:84% and the second longest wait times. 00:08:20.190 --> 00:08:22.150 align:middle line:84% Even though we have this free market system, 00:08:22.150 --> 00:08:24.930 align:middle line:84% we actually have the second longest wait times. 00:08:24.930 --> 00:08:30.170 align:middle line:84% So why is health care so expensive? 00:08:30.170 --> 00:08:33.270 align:middle line:84% Well, basically, if you look at figure 15-4, 00:08:33.270 --> 00:08:34.650 align:middle line:90% here is one estimate. 00:08:34.650 --> 00:08:40.890 align:middle line:84% This is an estimate of the extra essentially billion dollars. 00:08:40.890 --> 00:08:45.650 align:middle line:84% This is billion dollars as of 2018, 2019-- 00:08:45.650 --> 00:08:48.960 align:middle line:84% It's more now-- that it's viewed as excess medical spending. 00:08:48.960 --> 00:08:51.283 align:middle line:84% Now, obviously, this is a rough accounting exercise. 00:08:51.283 --> 00:08:53.700 align:middle line:84% You can't take every dollar and decide whether it's excess 00:08:53.700 --> 00:08:54.360 align:middle line:90% or not. 00:08:54.360 --> 00:08:57.180 align:middle line:84% But roughly speaking, this is trying to get at why are health 00:08:57.180 --> 00:08:59.540 align:middle line:84% so much more expensive than the rest of the world. 00:08:59.540 --> 00:09:00.360 align:middle line:90% So what's going on? 00:09:00.360 --> 00:09:02.340 align:middle line:84% Well, the first category in green 00:09:02.340 --> 00:09:05.640 align:middle line:90% is excess administrative costs. 00:09:05.640 --> 00:09:09.240 align:middle line:84% That's the easiest one to understand. 00:09:09.240 --> 00:09:11.220 align:middle line:84% The fact that it's easy to understand 00:09:11.220 --> 00:09:14.380 align:middle line:84% is basically our private health insurance system 00:09:14.380 --> 00:09:17.620 align:middle line:84% has much higher administrative costs than do public health care 00:09:17.620 --> 00:09:19.540 align:middle line:90% systems. 00:09:19.540 --> 00:09:21.780 align:middle line:84% Our health care system in the US, 00:09:21.780 --> 00:09:24.500 align:middle line:84% private health insurance has an administrative cost 00:09:24.500 --> 00:09:25.680 align:middle line:90% of about 15%. 00:09:25.680 --> 00:09:28.140 align:middle line:84% That is the difference between what-- remember, 00:09:28.140 --> 00:09:30.620 align:middle line:84% we solved our model of optimal insurance. 00:09:30.620 --> 00:09:32.700 align:middle line:90% We assumed a 0 profit insurer. 00:09:32.700 --> 00:09:35.100 align:middle line:84% That insurer paid out in benefits what 00:09:35.100 --> 00:09:36.780 align:middle line:90% they collected in premiums. 00:09:36.780 --> 00:09:39.700 align:middle line:84% In reality, in America, insurers collect about 15% 00:09:39.700 --> 00:09:42.580 align:middle line:84% more in premiums than they pay out in benefits. 00:09:42.580 --> 00:09:45.800 align:middle line:84% That's much, much higher than public insurance 00:09:45.800 --> 00:09:49.000 align:middle line:90% systems in other countries. 00:09:49.000 --> 00:09:51.900 align:middle line:84% Moreover, that's only one piece of our higher cost. 00:09:51.900 --> 00:09:54.000 align:middle line:84% That's the higher cost of health care insurance. 00:09:54.000 --> 00:09:56.400 align:middle line:84% There's also a higher cost of administering health care 00:09:56.400 --> 00:09:58.880 align:middle line:84% because we have this multi-payer system. 00:09:58.880 --> 00:10:00.660 align:middle line:84% And each payer has a different form. 00:10:00.660 --> 00:10:03.360 align:middle line:84% You've got to fill out all these forms and track things. 00:10:03.360 --> 00:10:06.600 align:middle line:84% The estimates on this are varied a lot. 00:10:06.600 --> 00:10:13.360 align:middle line:84% But some estimates suggest that as much as 1% of GDP in the US 00:10:13.360 --> 00:10:16.160 align:middle line:84% or 1.5% of GDP may just be wasted 00:10:16.160 --> 00:10:18.020 align:middle line:90% provider administrative costs. 00:10:18.020 --> 00:10:19.980 align:middle line:84% Now that's, I think probably an upper bound. 00:10:19.980 --> 00:10:21.480 align:middle line:84% But the bottom line is there's a lot 00:10:21.480 --> 00:10:25.440 align:middle line:84% of excess administrative costs, both in terms of what insurers 00:10:25.440 --> 00:10:28.560 align:middle line:84% spend and in terms of what providers have to spend to deal 00:10:28.560 --> 00:10:31.740 align:middle line:90% with our multi-payer system. 00:10:31.740 --> 00:10:38.040 align:middle line:84% Now, the next factor is we pay prices that are much higher. 00:10:38.040 --> 00:10:38.940 align:middle line:90% This says too high. 00:10:38.940 --> 00:10:41.680 align:middle line:84% But let's just start for now saying much higher 00:10:41.680 --> 00:10:43.187 align:middle line:90% than the rest of the world. 00:10:43.187 --> 00:10:45.270 align:middle line:84% We can see this with international comparison data 00:10:45.270 --> 00:10:45.810 align:middle line:90% again. 00:10:45.810 --> 00:10:49.233 align:middle line:90% Go to figure 15-5. 00:10:49.233 --> 00:10:50.650 align:middle line:84% So what we do here is we take data 00:10:50.650 --> 00:10:52.930 align:middle line:84% from a whole set of developed countries. 00:10:52.930 --> 00:10:54.450 align:middle line:84% And for a set of services, we graph 00:10:54.450 --> 00:10:57.930 align:middle line:84% the cost of the United States, the cost and the average 00:10:57.930 --> 00:11:01.130 align:middle line:84% of these developed countries, and the cost in the lowest cost 00:11:01.130 --> 00:11:02.650 align:middle line:90% country. 00:11:02.650 --> 00:11:05.210 align:middle line:84% What you see is in every case, the US 00:11:05.210 --> 00:11:09.290 align:middle line:84% is multiples of the average cost and many multiples 00:11:09.290 --> 00:11:10.070 align:middle line:90% of the low cost. 00:11:10.070 --> 00:11:12.528 align:middle line:84% So if you look at, for example, something simple like a hip 00:11:12.528 --> 00:11:15.850 align:middle line:84% replacement, which they do millions of all over the world 00:11:15.850 --> 00:11:20.650 align:middle line:84% every year, in the US, that costs on average about $32,000. 00:11:20.650 --> 00:11:23.790 align:middle line:84% This is all in comparable in US dollars. 00:11:23.790 --> 00:11:26.130 align:middle line:84% In the average developed country, it's about half that. 00:11:26.130 --> 00:11:31.730 align:middle line:84% In the lowest cost country, it's about a seventh or a sixth. 00:11:31.730 --> 00:11:35.230 align:middle line:84% You see the hip replacements, angioplasty, C-sections, 00:11:35.230 --> 00:11:37.050 align:middle line:90% delivery, appendectomies. 00:11:37.050 --> 00:11:39.870 align:middle line:84% And then you all see it for specific goods. 00:11:39.870 --> 00:11:41.290 align:middle line:84% What the second table does is show 00:11:41.290 --> 00:11:45.600 align:middle line:84% the cost difference we pay for prescription drugs in the US. 00:11:45.600 --> 00:11:47.860 align:middle line:84% Here are common brands of drugs, and we 00:11:47.860 --> 00:11:51.660 align:middle line:84% see is once again much higher prices paid in the US 00:11:51.660 --> 00:11:53.780 align:middle line:90% than in other countries. 00:11:53.780 --> 00:11:55.820 align:middle line:84% So the second reason we spend so much 00:11:55.820 --> 00:12:00.820 align:middle line:84% is that we pay a lot higher prices. 00:12:00.820 --> 00:12:02.660 align:middle line:84% Now, why are those prices higher? 00:12:02.660 --> 00:12:09.200 align:middle line:84% Well, let's ask ourselves what disciplines prices in a market. 00:12:09.200 --> 00:12:11.793 align:middle line:84% We have a market for health care. 00:12:11.793 --> 00:12:14.460 align:middle line:84% Unlike other countries, we don't have a single payer health care 00:12:14.460 --> 00:12:14.960 align:middle line:90% system. 00:12:14.960 --> 00:12:18.180 align:middle line:84% We have a free market for health for many of our payers. 00:12:18.180 --> 00:12:20.590 align:middle line:84% What disciplines prices in a free market? 00:12:20.590 --> 00:12:22.340 align:middle line:84% What keeps prices closer to marginal cost? 00:12:22.340 --> 00:12:22.720 align:middle line:90% Yeah? 00:12:22.720 --> 00:12:23.637 align:middle line:90% AUDIENCE: Competition. 00:12:23.637 --> 00:12:25.700 align:middle line:84% JON GRUBER: Competition and shopping. 00:12:25.700 --> 00:12:28.820 align:middle line:84% Well, both of those are problematic in health care. 00:12:28.820 --> 00:12:30.323 align:middle line:84% The competition part is problematic 00:12:30.323 --> 00:12:31.740 align:middle line:84% because there's an obvious feature 00:12:31.740 --> 00:12:34.680 align:middle line:84% of natural monopoly to many features of health care. 00:12:34.680 --> 00:12:37.380 align:middle line:84% If you have a heart attack on Nantucket, 00:12:37.380 --> 00:12:39.840 align:middle line:84% you have to go to the one hospital on Nantucket. 00:12:39.840 --> 00:12:42.542 align:middle line:84% You can't say, I'm going to go get me on a plane, 00:12:42.542 --> 00:12:44.250 align:middle line:84% put me over here a few hours, I'll be OK. 00:12:44.250 --> 00:12:45.670 align:middle line:90% You're not going to do that. 00:12:45.670 --> 00:12:47.833 align:middle line:84% Likewise, it's pretty hard to shop. 00:12:47.833 --> 00:12:50.250 align:middle line:84% If you're in the back of an ambulance with a heart attack, 00:12:50.250 --> 00:12:52.190 align:middle line:84% you're not going to be comparing prices. 00:12:52.190 --> 00:12:54.430 align:middle line:84% Moreover, if you wanted to, prices 00:12:54.430 --> 00:12:59.270 align:middle line:84% are incredibly opaque and hard to even figure out. 00:12:59.270 --> 00:13:01.830 align:middle line:84% Moreover, the market power here comes not just 00:13:01.830 --> 00:13:04.910 align:middle line:90% from geographic monopolies. 00:13:04.910 --> 00:13:07.750 align:middle line:84% It can come from also reputational monopolies. 00:13:07.750 --> 00:13:10.370 align:middle line:84% Hospitals can just be viewed as the best, 00:13:10.370 --> 00:13:12.230 align:middle line:84% regardless of whether they deserve it. 00:13:12.230 --> 00:13:15.190 align:middle line:84% When David Ortiz, beloved Red Sox player, 00:13:15.190 --> 00:13:16.990 align:middle line:84% was shot in the Dominican Republic 00:13:16.990 --> 00:13:19.510 align:middle line:84% and was flown back to the US for care, 00:13:19.510 --> 00:13:20.990 align:middle line:84% did he go to the official hospital 00:13:20.990 --> 00:13:23.270 align:middle line:84% of the Red Sox, one of the top 10 hospitals 00:13:23.270 --> 00:13:24.710 align:middle line:90% in the country, Beth Israel? 00:13:24.710 --> 00:13:27.330 align:middle line:84% No, he went to Mass General Hospital, 00:13:27.330 --> 00:13:29.430 align:middle line:90% which is rated somewhat higher. 00:13:29.430 --> 00:13:31.290 align:middle line:84% And they decide, well, Big Poppy's 00:13:31.290 --> 00:13:33.290 align:middle line:84% got to go to Mass General because that's better. 00:13:33.290 --> 00:13:35.010 align:middle line:84% Now, the truth is I can't say for his case 00:13:35.010 --> 00:13:36.930 align:middle line:84% whether it made a difference, but for most people, 00:13:36.930 --> 00:13:39.097 align:middle line:84% it doesn't make a difference if you go to MGH or BI, 00:13:39.097 --> 00:13:41.200 align:middle line:90% but MGH charges a lot more. 00:13:41.200 --> 00:13:44.760 align:middle line:84% That's a reputational monopoly, natural monopoly. 00:13:44.760 --> 00:13:47.040 align:middle line:84% There's basically imperfect competition 00:13:47.040 --> 00:13:51.640 align:middle line:84% is rife in the health care provider space. 00:13:51.640 --> 00:13:55.138 align:middle line:84% Indeed, the foundational article on health economics 00:13:55.138 --> 00:13:57.680 align:middle line:84% is written by Kenneth Arrow, a Nobel Prize-winning economist. 00:13:57.680 --> 00:13:59.305 align:middle line:84% Remember, Arrow's impossibility theorem 00:13:59.305 --> 00:14:01.040 align:middle line:90% we did earlier this semester. 00:14:01.040 --> 00:14:03.480 align:middle line:84% He wrote in 1963 an article which basically 00:14:03.480 --> 00:14:05.840 align:middle line:84% could be summarized as there's no market as fucked up 00:14:05.840 --> 00:14:08.480 align:middle line:84% as health care, basically said everything 00:14:08.480 --> 00:14:10.640 align:middle line:90% fails in health care in 1963. 00:14:10.640 --> 00:14:14.400 align:middle line:84% So we've known for a long time that basically market failures 00:14:14.400 --> 00:14:15.220 align:middle line:90% are rife here. 00:14:15.220 --> 00:14:18.560 align:middle line:90% 00:14:18.560 --> 00:14:24.400 align:middle line:90% Recognizing that, we find that-- 00:14:24.400 --> 00:14:26.600 align:middle line:84% recognizing that, what other countries do 00:14:26.600 --> 00:14:29.460 align:middle line:84% is regulate the prices that are charged for health care. 00:14:29.460 --> 00:14:32.640 align:middle line:84% Other countries say, well, we've learned in 14.01 00:14:32.640 --> 00:14:35.240 align:middle line:84% that if the market doesn't work, then there could be a welfare 00:14:35.240 --> 00:14:37.490 align:middle line:84% loss the government can correct, not necessarily 00:14:37.490 --> 00:14:39.170 align:middle line:90% will correct, but can correct. 00:14:39.170 --> 00:14:42.370 align:middle line:84% And every other country in the world regulates their prices. 00:14:42.370 --> 00:14:48.330 align:middle line:84% And that's why the prices are so much lower in other countries. 00:14:48.330 --> 00:14:53.317 align:middle line:84% If you look at figure 15-6, this shows how much variation we have 00:14:53.317 --> 00:14:54.150 align:middle line:90% in prices in the US. 00:14:54.150 --> 00:14:55.450 align:middle line:90% This is a pretty cool figure. 00:14:55.450 --> 00:14:59.570 align:middle line:84% This is basically across hospitals. 00:14:59.570 --> 00:15:03.770 align:middle line:84% Even for the same service, there is a massive variation 00:15:03.770 --> 00:15:06.990 align:middle line:90% in what's charged. 00:15:06.990 --> 00:15:08.390 align:middle line:90% Now some of this is geographic. 00:15:08.390 --> 00:15:09.790 align:middle line:90% This is for the whole nation. 00:15:09.790 --> 00:15:12.090 align:middle line:84% But even if you look within markets, 00:15:12.090 --> 00:15:15.470 align:middle line:84% there is things like 2 to 5 to 1 variation. 00:15:15.470 --> 00:15:17.450 align:middle line:84% Literally, the cost of basic service 00:15:17.450 --> 00:15:21.730 align:middle line:84% in the Boston medical market can vary 5 to 1 across hospitals. 00:15:21.730 --> 00:15:25.237 align:middle line:84% That is not the hallmark of a competitive market, 00:15:25.237 --> 00:15:27.070 align:middle line:84% That is not the hallmark competitive market. 00:15:27.070 --> 00:15:28.982 align:middle line:84% That basically even within the same market 00:15:28.982 --> 00:15:30.690 align:middle line:84% where an ambulance can pretty [INAUDIBLE] 00:15:30.690 --> 00:15:34.890 align:middle line:84% take you lots of places, there's enormous variation in prices. 00:15:34.890 --> 00:15:38.640 align:middle line:84% It's clearly shopping and competition are not working. 00:15:38.640 --> 00:15:42.005 align:middle line:90% 00:15:42.005 --> 00:15:45.700 align:middle line:84% And for example, in Tampa, Florida, 00:15:45.700 --> 00:15:48.860 align:middle line:84% if you look at the price of blood tests, there's a 40 to 1 00:15:48.860 --> 00:15:51.380 align:middle line:84% variation in the one city of Tampa in what it costs 00:15:51.380 --> 00:15:54.420 align:middle line:90% to get a blood test, 40 to 1. 00:15:54.420 --> 00:15:57.160 align:middle line:84% This is clearly competition and shopping are not working. 00:15:57.160 --> 00:15:59.750 align:middle line:84% The rest of the world is dealing with this through regulation. 00:15:59.750 --> 00:16:01.000 align:middle line:90% How can the US deal with that? 00:16:01.000 --> 00:16:04.880 align:middle line:84% And that's something we'll talk a lot about. 00:16:04.880 --> 00:16:10.660 align:middle line:84% Then if you go back to figure 15-4, 00:16:10.660 --> 00:16:13.700 align:middle line:84% they have inefficiently delivered and unnecessary 00:16:13.700 --> 00:16:14.240 align:middle line:90% services. 00:16:14.240 --> 00:16:17.560 align:middle line:84% Let's lump those together What does that mean? 00:16:17.560 --> 00:16:20.860 align:middle line:84% That means that we get a lot of health care in the US 00:16:20.860 --> 00:16:22.620 align:middle line:90% we do not need. 00:16:22.620 --> 00:16:25.200 align:middle line:84% And here the most famous example, 00:16:25.200 --> 00:16:28.500 align:middle line:84% and probably one of the most important health economics 00:16:28.500 --> 00:16:31.500 align:middle line:84% pieces ever written, was in 2009 article in The New Yorker 00:16:31.500 --> 00:16:33.760 align:middle line:90% magazine by Atul Gawande. 00:16:33.760 --> 00:16:37.280 align:middle line:84% He's both a physician and writer, brilliant guy, 00:16:37.280 --> 00:16:40.160 align:middle line:90% and huge live music fan. 00:16:40.160 --> 00:16:45.540 align:middle line:84% And basically, what he did is he compared two cities, McAllen, 00:16:45.540 --> 00:16:47.520 align:middle line:90% Texas and El Paso, Texas. 00:16:47.520 --> 00:16:51.120 align:middle line:84% These are two cities very near each other with very comparable 00:16:51.120 --> 00:16:53.640 align:middle line:84% population in terms of the age distribution, the income 00:16:53.640 --> 00:16:56.820 align:middle line:90% distribution, et cetera. 00:16:56.820 --> 00:16:59.880 align:middle line:84% And he looked at the Medicare program, which 00:16:59.880 --> 00:17:02.520 align:middle line:84% is I'll talk about next time, pays a fixed price 00:17:02.520 --> 00:17:03.720 align:middle line:90% for every service. 00:17:03.720 --> 00:17:05.037 align:middle line:90% So you got the same price. 00:17:05.037 --> 00:17:07.079 align:middle line:84% If you do an appendectomy in McAllen and El Paso, 00:17:07.079 --> 00:17:08.960 align:middle line:90% you get the same price. 00:17:08.960 --> 00:17:11.880 align:middle line:84% And yet Medicare spending per capita 00:17:11.880 --> 00:17:15.297 align:middle line:84% was twice as high in McAllen as El Paso. 00:17:15.297 --> 00:17:17.339 align:middle line:84% Ignoring the price variation I just talked about, 00:17:17.339 --> 00:17:19.760 align:middle line:84% this is purely intensity variation. 00:17:19.760 --> 00:17:25.040 align:middle line:84% So he dug in, and what he found is in McAllen, 00:17:25.040 --> 00:17:28.040 align:middle line:84% patients got 50% more specialist visits. 00:17:28.040 --> 00:17:31.360 align:middle line:84% They were 2/3 more likely to see 10 or more specialists. 00:17:31.360 --> 00:17:35.090 align:middle line:84% They received 20% more abdominal ultrasounds, 60% more 00:17:35.090 --> 00:17:38.090 align:middle line:84% stress tests, 200% more nerve conduction studies 00:17:38.090 --> 00:17:42.970 align:middle line:84% to diagnose carpal tunnel, and 550% more urine flow studies 00:17:42.970 --> 00:17:47.930 align:middle line:84% to diagnose prostate troubles, 2 to 3 times as many pacemakers 00:17:47.930 --> 00:17:51.130 align:middle line:84% and 5 times as many home nurse visits. 00:17:51.130 --> 00:17:52.530 align:middle line:90% So what's going on? 00:17:52.530 --> 00:17:55.530 align:middle line:84% Well, basically what Gawande argued 00:17:55.530 --> 00:17:59.050 align:middle line:84% is that essentially we got into an equilibrium 00:17:59.050 --> 00:18:02.450 align:middle line:84% where the doctors in McAllen were greedy 00:18:02.450 --> 00:18:04.270 align:middle line:84% and the doctors in El Paso were less so. 00:18:04.270 --> 00:18:06.790 align:middle line:84% Doctors in McAllen realized that the more they did, 00:18:06.790 --> 00:18:08.450 align:middle line:84% the more money they made, and they just 00:18:08.450 --> 00:18:10.950 align:middle line:84% went off the reservation, just started doing a ton of stuff, 00:18:10.950 --> 00:18:14.210 align:middle line:84% became the culture of McAllen was to overtreat people. 00:18:14.210 --> 00:18:18.130 align:middle line:84% And he said overtreating because there's a fascinating fact. 00:18:18.130 --> 00:18:21.450 align:middle line:84% Despite spending twice as much per capita, 00:18:21.450 --> 00:18:25.030 align:middle line:84% people live no longer in McAllen. 00:18:25.030 --> 00:18:26.710 align:middle line:84% There was no extra length of life, 00:18:26.710 --> 00:18:28.650 align:middle line:84% no obvious improvement in health, 00:18:28.650 --> 00:18:31.460 align:middle line:90% despite spending twice as much. 00:18:31.460 --> 00:18:33.460 align:middle line:84% And that's what he meant by overtreatment, 00:18:33.460 --> 00:18:36.140 align:middle line:84% that basically we're spending doing all this extra stuff 00:18:36.140 --> 00:18:38.220 align:middle line:84% and people aren't living any longer. 00:18:38.220 --> 00:18:41.100 align:middle line:84% And indeed, Gawande gets all the credit for this, but, actually, 00:18:41.100 --> 00:18:43.660 align:middle line:84% what he did is popularize a resource that economists had 00:18:43.660 --> 00:18:46.380 align:middle line:84% known about called the Dartmouth Health Atlas, 00:18:46.380 --> 00:18:49.180 align:middle line:84% which was created in the 2000s, essentially took Medicare data 00:18:49.180 --> 00:18:51.900 align:middle line:84% and showed for every area in the country what were outcomes 00:18:51.900 --> 00:18:53.440 align:middle line:90% and what did health care cost. 00:18:53.440 --> 00:18:55.300 align:middle line:90% And there's no correlation. 00:18:55.300 --> 00:18:58.040 align:middle line:84% Health care costs vary enormously around the country, 00:18:58.040 --> 00:19:00.880 align:middle line:84% with no correlation with outcomes, suggesting-- 00:19:00.880 --> 00:19:03.300 align:middle line:84% and that's for Medicare where prices are fixed. 00:19:03.300 --> 00:19:06.000 align:middle line:84% So suggesting that in addition to excess prices, 00:19:06.000 --> 00:19:09.660 align:middle line:84% we're paying for a lot of inefficient care. 00:19:09.660 --> 00:19:14.095 align:middle line:90% 00:19:14.095 --> 00:19:15.800 align:middle line:84% And that's not just that evidence. 00:19:15.800 --> 00:19:18.180 align:middle line:84% There's lots of evidence from things like chart reviews 00:19:18.180 --> 00:19:21.140 align:middle line:84% where doctors go back and look at charts of patients treatment 00:19:21.140 --> 00:19:24.700 align:middle line:84% and show how they were excessively treated. 00:19:24.700 --> 00:19:27.900 align:middle line:84% For example, by the early 2000s, there were four different 00:19:27.900 --> 00:19:30.670 align:middle line:84% clinical trials showing that showing spinal fusion surgery 00:19:30.670 --> 00:19:32.490 align:middle line:84% was no better than non-surgical treatment. 00:19:32.490 --> 00:19:34.870 align:middle line:84% So one of the most overused procedures in medicine, 00:19:34.870 --> 00:19:36.912 align:middle line:84% spinal fusion surgery-- when you have a bad back, 00:19:36.912 --> 00:19:38.910 align:middle line:90% they fuse your spine. 00:19:38.910 --> 00:19:41.010 align:middle line:84% Despite this evidence, back surgery rates 00:19:41.010 --> 00:19:42.510 align:middle line:84% are high and growing in the US, even 00:19:42.510 --> 00:19:44.230 align:middle line:84% though medical evidence consistently 00:19:44.230 --> 00:19:45.570 align:middle line:90% says we shouldn't be doing it. 00:19:45.570 --> 00:19:48.670 align:middle line:90% 00:19:48.670 --> 00:19:54.170 align:middle line:84% Here's another great example, oral intravenous acetaminophen, 00:19:54.170 --> 00:19:58.910 align:middle line:84% so Tylenol through an IV, costs hundreds of times 00:19:58.910 --> 00:20:02.230 align:middle line:90% as much as oral acetaminophen. 00:20:02.230 --> 00:20:04.170 align:middle line:84% There's no evidence it works any better. 00:20:04.170 --> 00:20:05.130 align:middle line:84% Obviously, if you're unconscious, 00:20:05.130 --> 00:20:05.690 align:middle line:90% there's a difference. 00:20:05.690 --> 00:20:08.430 align:middle line:84% But if you're conscious, there's no evidence it works any better. 00:20:08.430 --> 00:20:11.390 align:middle line:84% And yet intravenous acetaminophen 00:20:11.390 --> 00:20:14.190 align:middle line:84% is used all over the country by many hospitals at a cost 00:20:14.190 --> 00:20:16.270 align:middle line:90% hundreds of times as high. 00:20:16.270 --> 00:20:20.110 align:middle line:84% So basically, we have this expensive system 00:20:20.110 --> 00:20:22.510 align:middle line:84% because we have a lot of administrative waste. 00:20:22.510 --> 00:20:24.150 align:middle line:84% We have prices that are much higher, 00:20:24.150 --> 00:20:26.738 align:middle line:90% and we do a lot of extra care. 00:20:26.738 --> 00:20:28.280 align:middle line:84% Now there's also another bullet here. 00:20:28.280 --> 00:20:32.400 align:middle line:84% If you go back 15-4 for fraud, care coordination failures 00:20:32.400 --> 00:20:33.960 align:middle line:84% probably could be lumped in also. 00:20:33.960 --> 00:20:36.640 align:middle line:84% That's people not getting passed correctly through their doctors 00:20:36.640 --> 00:20:36.760 align:middle line:90% that. 00:20:36.760 --> 00:20:39.000 align:middle line:84% Could also be lumped in with unnecessary service, 00:20:39.000 --> 00:20:40.440 align:middle line:90% inefficient services. 00:20:40.440 --> 00:20:42.200 align:middle line:90% Fraud is interesting. 00:20:42.200 --> 00:20:45.680 align:middle line:84% Once again, we have a lot of health care fraud. 00:20:45.680 --> 00:20:48.720 align:middle line:84% So we have lots of situations where people are literally 00:20:48.720 --> 00:20:51.400 align:middle line:84% fraudulently billing for services they didn't perform, 00:20:51.400 --> 00:20:54.840 align:middle line:84% et cetera, once again, a problem that 00:20:54.840 --> 00:20:57.240 align:middle line:84% appears to be worse in the US, not because we're 00:20:57.240 --> 00:20:59.860 align:middle line:90% necessarily more dishonest. 00:20:59.860 --> 00:21:01.820 align:middle line:84% It's not the US is necessarily more dishonest. 00:21:01.820 --> 00:21:04.360 align:middle line:84% It's that there's so much more money to be made through it. 00:21:04.360 --> 00:21:07.685 align:middle line:84% If prices are low, it's not as valuable to commit fraud. 00:21:07.685 --> 00:21:09.060 align:middle line:84% When you can make a ton of money, 00:21:09.060 --> 00:21:10.420 align:middle line:90% fraud becomes more valuable. 00:21:10.420 --> 00:21:13.640 align:middle line:84% And that's why we have a lot more of it. 00:21:13.640 --> 00:21:15.540 align:middle line:84% So that's where we are in health care costs. 00:21:15.540 --> 00:21:16.120 align:middle line:90% Yeah, Steven? 00:21:16.120 --> 00:21:18.900 align:middle line:84% AUDIENCE: [INAUDIBLE] for example, 00:21:18.900 --> 00:21:23.880 align:middle line:84% [INAUDIBLE] Tylenol instead of taking it orally, 00:21:23.880 --> 00:21:30.690 align:middle line:84% is that the health care providers are getting some bonus 00:21:30.690 --> 00:21:33.690 align:middle line:84% based on what is it they give, or is it just strictly 00:21:33.690 --> 00:21:36.250 align:middle line:84% they're not paying the bill because of insurance so they 00:21:36.250 --> 00:21:37.090 align:middle line:90% just don't-- 00:21:37.090 --> 00:21:37.670 align:middle line:90% JON GRUBER: Great question. 00:21:37.670 --> 00:21:38.230 align:middle line:90% And we'll cover that. 00:21:38.230 --> 00:21:40.290 align:middle line:84% We'll spend a lot of time on that this lecture. 00:21:40.290 --> 00:21:41.023 align:middle line:90% Yeah? 00:21:41.023 --> 00:21:43.190 align:middle line:84% AUDIENCE: We mentioned like reputational monopolies. 00:21:43.190 --> 00:21:45.315 align:middle line:84% I also know that a lot of times like, I don't know, 00:21:45.315 --> 00:21:47.870 align:middle line:84% insurance companies, and hospitals, 00:21:47.870 --> 00:21:49.890 align:middle line:84% and I guess care providers are all either 00:21:49.890 --> 00:21:54.330 align:middle line:84% owned by the same company or basically [INAUDIBLE] company? 00:21:54.330 --> 00:21:56.050 align:middle line:90% Would that also be [INAUDIBLE]? 00:21:56.050 --> 00:21:58.370 align:middle line:84% JON GRUBER: Yeah, I mean, basically this is a big-- 00:21:58.370 --> 00:22:02.470 align:middle line:84% and we'll come and talk about this some. 00:22:02.470 --> 00:22:05.450 align:middle line:84% But basically, consolidation in the health care industry 00:22:05.450 --> 00:22:06.630 align:middle line:90% has been massive. 00:22:06.630 --> 00:22:11.150 align:middle line:84% I think the view is it's been a bad thing for consumers. 00:22:11.150 --> 00:22:11.850 align:middle line:90% Yeah? 00:22:11.850 --> 00:22:15.250 align:middle line:84% AUDIENCE: Would you say that regarding drugs, the FDA's 00:22:15.250 --> 00:22:19.890 align:middle line:84% rigid patent and approval system either helps the market 00:22:19.890 --> 00:22:22.370 align:middle line:84% by like addressing research positive externalities 00:22:22.370 --> 00:22:25.555 align:middle line:84% or hurts it by pushing prices really, really high? 00:22:25.555 --> 00:22:26.680 align:middle line:90% JON GRUBER: Great question. 00:22:26.680 --> 00:22:27.780 align:middle line:84% And this is the kind of stuff I want 00:22:27.780 --> 00:22:28.860 align:middle line:84% to be-- you guys have a lot of questions 00:22:28.860 --> 00:22:29.960 align:middle line:90% about health care in your head. 00:22:29.960 --> 00:22:30.460 align:middle line:90% I know you do. 00:22:30.460 --> 00:22:31.120 align:middle line:90% Let's discuss them. 00:22:31.120 --> 00:22:32.120 align:middle line:90% So let's talk about the FDA. 00:22:32.120 --> 00:22:33.480 align:middle line:84% This is one I wouldn't come back to normally. 00:22:33.480 --> 00:22:35.220 align:middle line:84% So I'll spend a few minutes on this one. 00:22:35.220 --> 00:22:37.800 align:middle line:84% So the FDA, as we learned in chapter 1, 00:22:37.800 --> 00:22:40.470 align:middle line:84% regulates all the food we eat and all the drugs we take, 00:22:40.470 --> 00:22:42.420 align:middle line:90% et cetera. 00:22:42.420 --> 00:22:44.700 align:middle line:84% There's a deep question about how high 00:22:44.700 --> 00:22:47.500 align:middle line:84% should the hurdle be to approve a drug. 00:22:47.500 --> 00:22:50.100 align:middle line:84% The FDA has what's considered internationally 00:22:50.100 --> 00:22:55.300 align:middle line:84% a fairly high hurdle, usually, although sometimes lower, 00:22:55.300 --> 00:22:59.020 align:middle line:90% but traditionally higher. 00:22:59.020 --> 00:23:01.740 align:middle line:84% Here's what's interesting about that. 00:23:01.740 --> 00:23:03.700 align:middle line:84% I'm not qualified to say if it's too high. 00:23:03.700 --> 00:23:06.540 align:middle line:84% But here's what I do know, which is a fascinating disparity 00:23:06.540 --> 00:23:08.980 align:middle line:84% in the US, which is to get a drug approved, 00:23:08.980 --> 00:23:10.520 align:middle line:90% there's this very high barrier. 00:23:10.520 --> 00:23:12.220 align:middle line:90% Once it's approved, go to town. 00:23:12.220 --> 00:23:15.125 align:middle line:90% We don't even watch you anymore. 00:23:15.125 --> 00:23:17.500 align:middle line:84% And only when someone dies and if the people start to sue 00:23:17.500 --> 00:23:18.820 align:middle line:90% you do we pay attention. 00:23:18.820 --> 00:23:20.140 align:middle line:90% That's nuts. 00:23:20.140 --> 00:23:21.980 align:middle line:84% OK, it's crazy that it's incredibly hard 00:23:21.980 --> 00:23:23.140 align:middle line:90% to get it on the market. 00:23:23.140 --> 00:23:26.060 align:middle line:84% Then once it's on the market, we don't do much to keep it-- 00:23:26.060 --> 00:23:26.860 align:middle line:90% I'm overstating it. 00:23:26.860 --> 00:23:27.820 align:middle line:90% We do keep tabs on you. 00:23:27.820 --> 00:23:29.403 align:middle line:84% And if there's a lot of bad incidents, 00:23:29.403 --> 00:23:31.200 align:middle line:84% the FDA collects the data and investigates. 00:23:31.200 --> 00:23:33.000 align:middle line:84% But the truth is, we're a lot tougher 00:23:33.000 --> 00:23:34.540 align:middle line:84% about approving the [INAUDIBLE] that are on the market. 00:23:34.540 --> 00:23:36.915 align:middle line:84% That disparity doesn't seem to make a lot of sense to me. 00:23:36.915 --> 00:23:39.080 align:middle line:90% 00:23:39.080 --> 00:23:40.300 align:middle line:90% Yeah, Steven? 00:23:40.300 --> 00:23:41.300 align:middle line:90% AUDIENCE: Spinal fusion. 00:23:41.300 --> 00:23:43.380 align:middle line:90% Isn't there also an issue of-- 00:23:43.380 --> 00:23:46.400 align:middle line:84% like that surgery makes people-- like it 00:23:46.400 --> 00:23:47.780 align:middle line:90% reduces people's mobility. 00:23:47.780 --> 00:23:49.920 align:middle line:84% JON GRUBER: Well, I mean, that's a separate thing 00:23:49.920 --> 00:23:52.440 align:middle line:84% of whether you're actually even worse off getting 00:23:52.440 --> 00:23:53.580 align:middle line:90% that surgery than not. 00:23:53.580 --> 00:23:54.960 align:middle line:90% I'm not even going there. 00:23:54.960 --> 00:23:57.140 align:middle line:84% But clearly, it's not recommended medically. 00:23:57.140 --> 00:23:59.280 align:middle line:90% And yet it's done excessively. 00:23:59.280 --> 00:24:00.160 align:middle line:90% Yeah? 00:24:00.160 --> 00:24:02.680 align:middle line:84% AUDIENCE: What about generic versus brand name 00:24:02.680 --> 00:24:04.372 align:middle line:90% like medications? 00:24:04.372 --> 00:24:05.080 align:middle line:90% JON GRUBER: Yeah. 00:24:05.080 --> 00:24:07.200 align:middle line:84% AUDIENCE: Is that another example of [INAUDIBLE]? 00:24:07.200 --> 00:24:10.360 align:middle line:84% JON GRUBER: Yeah, I mean, look, we talked in 14.01 00:24:10.360 --> 00:24:11.560 align:middle line:90% about the role of patents. 00:24:11.560 --> 00:24:14.080 align:middle line:84% And basically, you remember 14.01, 00:24:14.080 --> 00:24:16.463 align:middle line:84% we talked about where monopolies come from. 00:24:16.463 --> 00:24:18.380 align:middle line:84% And one place monopolies come from is patents. 00:24:18.380 --> 00:24:20.240 align:middle line:84% And there's a trade-off with patents. 00:24:20.240 --> 00:24:22.610 align:middle line:84% The trade-off is, on the one hand, 00:24:22.610 --> 00:24:26.370 align:middle line:84% you want to create rents to incentivize innovation. 00:24:26.370 --> 00:24:29.670 align:middle line:84% On the other hand, once the good's innovated and produced, 00:24:29.670 --> 00:24:30.930 align:middle line:90% there's a monopoly. 00:24:30.930 --> 00:24:34.450 align:middle line:84% And the optimal patent structure would trade that off. 00:24:34.450 --> 00:24:37.730 align:middle line:84% In a perfect world, we would do what the Nobel Prize 00:24:37.730 --> 00:24:39.930 align:middle line:84% winning economist Michael Kremer suggested, 00:24:39.930 --> 00:24:42.170 align:middle line:84% which is society would decide what an innovation is 00:24:42.170 --> 00:24:44.510 align:middle line:84% worth and would pay you a price for inventing it 00:24:44.510 --> 00:24:46.507 align:middle line:84% and then sell it at marginal cost. 00:24:46.507 --> 00:24:48.590 align:middle line:84% Then that would be the most efficient thing to do. 00:24:48.590 --> 00:24:49.713 align:middle line:90% We don't do that. 00:24:49.713 --> 00:24:52.130 align:middle line:84% So what we do is we're going to set up this patent system. 00:24:52.130 --> 00:24:55.410 align:middle line:84% And are patents too long or too short or too tough? 00:24:55.410 --> 00:24:57.590 align:middle line:84% That is a fascinating and difficult area. 00:24:57.590 --> 00:25:00.450 align:middle line:84% Let me mention one area which illustrates 00:25:00.450 --> 00:25:01.910 align:middle line:90% the challenge with patents. 00:25:01.910 --> 00:25:04.630 align:middle line:84% This is by a famous article by Heidi Williams, 00:25:04.630 --> 00:25:06.890 align:middle line:90% who's at Dartmouth College. 00:25:06.890 --> 00:25:12.090 align:middle line:84% Heidi showed that if you look at drugs and the amount of money 00:25:12.090 --> 00:25:17.330 align:middle line:84% that's invested by the private sector, the shorter the trial, 00:25:17.330 --> 00:25:19.170 align:middle line:84% the more the private sector will invest. 00:25:19.170 --> 00:25:21.100 align:middle line:84% The private sector massively underinvests 00:25:21.100 --> 00:25:22.642 align:middle line:84% in drugs where it takes like 10 years 00:25:22.642 --> 00:25:23.900 align:middle line:90% to figure out if they work. 00:25:23.900 --> 00:25:24.660 align:middle line:90% Why? 00:25:24.660 --> 00:25:26.780 align:middle line:84% Because the patent clock starts ticking the minute 00:25:26.780 --> 00:25:27.762 align:middle line:90% you file the patent. 00:25:27.762 --> 00:25:29.720 align:middle line:84% So it takes 10 years to figure out if it works, 00:25:29.720 --> 00:25:32.500 align:middle line:84% you only have a few years of patent left before you 00:25:32.500 --> 00:25:34.180 align:middle line:90% lose your monopoly rights. 00:25:34.180 --> 00:25:36.380 align:middle line:84% So that's the kind of thing where patents clearly 00:25:36.380 --> 00:25:38.240 align:middle line:90% affect incentives of companies. 00:25:38.240 --> 00:25:40.620 align:middle line:84% And there's a whole separate, fascinating field 00:25:40.620 --> 00:25:42.460 align:middle line:84% of patent economics and how do we 00:25:42.460 --> 00:25:44.960 align:middle line:84% think about the optimal patent, how it should be structured. 00:25:44.960 --> 00:25:46.140 align:middle line:90% It's a great question. 00:25:46.140 --> 00:25:47.080 align:middle line:90% Yeah? 00:25:47.080 --> 00:25:52.620 align:middle line:84% AUDIENCE: [INAUDIBLE] but that was the one 00:25:52.620 --> 00:25:54.740 align:middle line:84% that you mentioned at the beginning of the course 00:25:54.740 --> 00:25:55.737 align:middle line:90% of the COVID vaccine. 00:25:55.737 --> 00:25:57.820 align:middle line:84% JON GRUBER: Yeah, I mean, basically, COVID vaccine 00:25:57.820 --> 00:26:00.720 align:middle line:84% is one example where we basically put in a prize. 00:26:00.720 --> 00:26:04.230 align:middle line:84% The prize in that case was a government guaranteed purchase. 00:26:04.230 --> 00:26:10.300 align:middle line:84% AUDIENCE: [INAUDIBLE] if this was a subsidy [INAUDIBLE]. 00:26:10.300 --> 00:26:11.700 align:middle line:90% JON GRUBER: No. 00:26:11.700 --> 00:26:13.680 align:middle line:90% Basically, it was a subsidy. 00:26:13.680 --> 00:26:16.540 align:middle line:84% It was essentially what we call a pull incentive, saying 00:26:16.540 --> 00:26:18.750 align:middle line:84% basically, look, if you invent a good vaccine, 00:26:18.750 --> 00:26:22.215 align:middle line:84% the prize is we'll guarantee we'll buy it. 00:26:22.215 --> 00:26:24.090 align:middle line:84% It wasn't just a pure prize for inventing it. 00:26:24.090 --> 00:26:25.327 align:middle line:90% You had to then deliver it. 00:26:25.327 --> 00:26:26.910 align:middle line:84% But since marginal cost was really low 00:26:26.910 --> 00:26:30.310 align:middle line:84% compared to fixed costs, we're essentially going to pay you, 00:26:30.310 --> 00:26:33.230 align:middle line:84% guarantee you'll cover your fixed costs if you invent it. 00:26:33.230 --> 00:26:33.770 align:middle line:90% Yeah? 00:26:33.770 --> 00:26:41.110 align:middle line:84% AUDIENCE: [INAUDIBLE] we already have [INAUDIBLE] Medicare. 00:26:41.110 --> 00:26:46.150 align:middle line:84% Would it be a helpful framework to have more medical innovations 00:26:46.150 --> 00:26:48.110 align:middle line:84% sponsored in the way that the COVID vaccine 00:26:48.110 --> 00:26:51.250 align:middle line:84% was where you have the prize system that you were mentioning, 00:26:51.250 --> 00:26:54.513 align:middle line:84% but the prize is like Medicare is already [INAUDIBLE]. 00:26:54.513 --> 00:26:55.930 align:middle line:84% JON GRUBER: It's a great question. 00:26:55.930 --> 00:26:58.450 align:middle line:84% I mean, this is something that worked very well in COVID. 00:26:58.450 --> 00:27:00.910 align:middle line:84% I think the bottom line on that is 00:27:00.910 --> 00:27:03.190 align:middle line:84% you'd want to essentially do that where 00:27:03.190 --> 00:27:07.130 align:middle line:84% you think the private market is least likely to deliver. 00:27:07.130 --> 00:27:09.295 align:middle line:84% So, for example, classic case is vaccines. 00:27:09.295 --> 00:27:11.670 align:middle line:84% The private market-- I think I covered this-- underinvest 00:27:11.670 --> 00:27:13.170 align:middle line:84% in vaccines because they're not sure 00:27:13.170 --> 00:27:16.740 align:middle line:84% if the thing will get cured, if it'll go away or not, 00:27:16.740 --> 00:27:21.095 align:middle line:84% vaccines or another one is antimicrobial resistance 00:27:21.095 --> 00:27:23.720 align:middle line:84% that basically there's all these antibiotics that we're getting 00:27:23.720 --> 00:27:27.053 align:middle line:84% [INAUDIBLE] and no one's invested in a new cure to that. 00:27:27.053 --> 00:27:29.220 align:middle line:84% You wouldn't want to say the government does it all, 00:27:29.220 --> 00:27:31.840 align:middle line:84% but I think targeting things like that, where clearly there's 00:27:31.840 --> 00:27:33.480 align:middle line:84% the private incentives are too low, 00:27:33.480 --> 00:27:36.000 align:middle line:84% would be the natural place to go. 00:27:36.000 --> 00:27:38.917 align:middle line:84% So that's the basic of health care spending. 00:27:38.917 --> 00:27:40.500 align:middle line:84% Let's now talk about health insurance. 00:27:40.500 --> 00:27:41.920 align:middle line:84% Much of our discussion in the next few lectures 00:27:41.920 --> 00:27:43.600 align:middle line:90% will focus on health insurance. 00:27:43.600 --> 00:27:46.660 align:middle line:84% So let's talk about how health insurance works. 00:27:46.660 --> 00:27:49.920 align:middle line:84% Fundamentally, health insurance is a deal 00:27:49.920 --> 00:27:54.320 align:middle line:84% where you pay a regular payment, typically monthly or bi-monthly, 00:27:54.320 --> 00:27:58.280 align:middle line:84% which is called the premium, in return for which the health 00:27:58.280 --> 00:28:03.040 align:middle line:84% insurer covers the cost of your medical care minus 00:28:03.040 --> 00:28:06.520 align:middle line:90% what's called some cost sharing. 00:28:06.520 --> 00:28:11.720 align:middle line:84% So at MIT, I pay my I pay my bi-monthly premium. 00:28:11.720 --> 00:28:14.060 align:middle line:84% The health insurer pays my health care costs, 00:28:14.060 --> 00:28:16.430 align:middle line:84% but I have to pay some of it too. 00:28:16.430 --> 00:28:17.477 align:middle line:90% It's called cost sharing. 00:28:17.477 --> 00:28:19.310 align:middle line:84% That cost sharing can come in several forms. 00:28:19.310 --> 00:28:21.450 align:middle line:84% The most common form is a deductible. 00:28:21.450 --> 00:28:23.470 align:middle line:84% So at MIT, I'm in a high deductible plan. 00:28:23.470 --> 00:28:26.730 align:middle line:84% What that means is the first $2,500 of my family's health 00:28:26.730 --> 00:28:28.810 align:middle line:90% care spending every year, I pay. 00:28:28.810 --> 00:28:31.273 align:middle line:90% Then the insurer kicks in. 00:28:31.273 --> 00:28:32.690 align:middle line:84% An alternative is what some of you 00:28:32.690 --> 00:28:33.930 align:middle line:84% may have dealt with a lot the doctor, which 00:28:33.930 --> 00:28:36.490 align:middle line:84% is a co-payment, which is you pay $20 every time you 00:28:36.490 --> 00:28:38.410 align:middle line:90% go to the doctor. 00:28:38.410 --> 00:28:40.858 align:middle line:84% Another alternative is what's called co-insurance, 00:28:40.858 --> 00:28:42.650 align:middle line:84% which is you pay X percent of the bill when 00:28:42.650 --> 00:28:44.210 align:middle line:90% you go to the doctor. 00:28:44.210 --> 00:28:47.170 align:middle line:84% All of these are different forms of cost sharing. 00:28:47.170 --> 00:28:49.970 align:middle line:84% So the key feature of an insurance contract 00:28:49.970 --> 00:28:52.050 align:middle line:90% is what does it cover? 00:28:52.050 --> 00:28:53.530 align:middle line:84% Under the ACA, insurance contracts 00:28:53.530 --> 00:28:55.290 align:middle line:84% cover pretty comparable set of things 00:28:55.290 --> 00:28:58.450 align:middle line:84% because we mandated a set of benefits they have to cover. 00:28:58.450 --> 00:29:01.250 align:middle line:84% So probably the two big distinctions 00:29:01.250 --> 00:29:03.650 align:middle line:84% across insurance products in the US 00:29:03.650 --> 00:29:08.290 align:middle line:84% are, A, how much cost sharing you have and, B, which doctors 00:29:08.290 --> 00:29:09.655 align:middle line:90% they include in their network. 00:29:09.655 --> 00:29:12.030 align:middle line:84% And that comes to the point your team was making earlier. 00:29:12.030 --> 00:29:13.530 align:middle line:84% We'll come back to that in a second. 00:29:13.530 --> 00:29:16.190 align:middle line:84% So let's focus on the first issue, which is essentially 00:29:16.190 --> 00:29:18.670 align:middle line:90% how much people have to pay. 00:29:18.670 --> 00:29:22.250 align:middle line:84% But first, let's talk about health insurance in the US. 00:29:22.250 --> 00:29:24.510 align:middle line:84% Table 15-1 shows the distribution of health coverage, 00:29:24.510 --> 00:29:25.593 align:middle line:90% just a little out of date. 00:29:25.593 --> 00:29:27.190 align:middle line:90% This is now 2020. 00:29:27.190 --> 00:29:28.450 align:middle line:90% It's not that different today. 00:29:28.450 --> 00:29:31.830 align:middle line:84% Essentially, we have, three quarters 00:29:31.830 --> 00:29:34.510 align:middle line:84% of the population or 70% of the population 00:29:34.510 --> 00:29:37.790 align:middle line:90% covered by private insurance. 00:29:37.790 --> 00:29:40.870 align:middle line:84% Almost all of that comes through employers. 00:29:40.870 --> 00:29:44.690 align:middle line:84% Most private insurers come through employers. 00:29:44.690 --> 00:29:47.510 align:middle line:84% But a substantial minority comes through direct purchase, 00:29:47.510 --> 00:29:50.670 align:middle line:84% through what we call the non-group insurance market. 00:29:50.670 --> 00:29:54.618 align:middle line:84% So employer insurance is typically of the form-- 00:29:54.618 --> 00:29:56.410 align:middle line:84% so I can just give you the numbers for MIT. 00:29:56.410 --> 00:30:03.115 align:middle line:84% MIT spends about 24 grand a year on my health insurance, 00:30:03.115 --> 00:30:04.490 align:middle line:84% but they don't spend all of that. 00:30:04.490 --> 00:30:06.823 align:middle line:84% They buy a health insurance plan for me from Blue Cross, 00:30:06.823 --> 00:30:08.030 align:middle line:90% cost $24,000 a year. 00:30:08.030 --> 00:30:10.570 align:middle line:84% They then asked me to pay some of the premiums. 00:30:10.570 --> 00:30:14.080 align:middle line:84% So the premiums are paid partly by my employer and partly by me. 00:30:14.080 --> 00:30:16.420 align:middle line:84% So typical arrangement is your employer will say, 00:30:16.420 --> 00:30:17.420 align:middle line:84% we're going to get you health insurance, 00:30:17.420 --> 00:30:19.500 align:middle line:84% but you have to pick up some of the premiums. 00:30:19.500 --> 00:30:21.637 align:middle line:84% So I probably pay about 25% of the premiums. 00:30:21.637 --> 00:30:23.220 align:middle line:84% I probably pay about six grand a year. 00:30:23.220 --> 00:30:25.000 align:middle line:90% MIT pays 18 grand. 00:30:25.000 --> 00:30:29.040 align:middle line:84% So that is the structure of a typical employer insurance 00:30:29.040 --> 00:30:31.620 align:middle line:84% plan will work is they'll buy the insurance. 00:30:31.620 --> 00:30:34.220 align:middle line:84% And then typically for a family plan, 00:30:34.220 --> 00:30:36.800 align:middle line:84% the employee pays about a quarter of the cost. 00:30:36.800 --> 00:30:38.878 align:middle line:84% OK, a non-group insurance plan is 00:30:38.878 --> 00:30:40.920 align:middle line:84% one where you buy it and you pay it all yourself. 00:30:40.920 --> 00:30:42.840 align:middle line:90% The employer is not involved. 00:30:42.840 --> 00:30:46.600 align:middle line:84% The non-group insurance market was a market 00:30:46.600 --> 00:30:49.518 align:middle line:84% that was fundamentally broken before the Affordable Care Act. 00:30:49.518 --> 00:30:51.560 align:middle line:84% And I'll talk about that, how the Affordable Care 00:30:51.560 --> 00:30:52.340 align:middle line:90% Act fixed it. 00:30:52.340 --> 00:30:55.040 align:middle line:84% But now it's a very functional, comparable market 00:30:55.040 --> 00:30:56.440 align:middle line:90% to the employer market. 00:30:56.440 --> 00:31:00.160 align:middle line:84% So the question is if you're going to buy private insurance, 00:31:00.160 --> 00:31:02.580 align:middle line:84% why do you buy through the employer market? 00:31:02.580 --> 00:31:06.500 align:middle line:84% Why is the employer market four times as big as the-- 00:31:06.500 --> 00:31:12.630 align:middle line:84% why is the employer market, yeah, four to five times as big? 00:31:12.630 --> 00:31:17.360 align:middle line:84% No, 6 times as big as the direct purchase market. 00:31:17.360 --> 00:31:19.110 align:middle line:84% Why is the employer market so much bigger? 00:31:19.110 --> 00:31:19.705 align:middle line:90% Yeah? 00:31:19.705 --> 00:31:20.580 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:31:20.580 --> 00:31:21.830 align:middle line:90% JON GRUBER: Well, OK. 00:31:21.830 --> 00:31:22.330 align:middle line:90% Go ahead. 00:31:22.330 --> 00:31:24.142 align:middle line:84% AUDIENCE: [INAUDIBLE] income tax? 00:31:24.142 --> 00:31:26.120 align:middle line:84% JON GRUBER: Well, OK, that's subtle. 00:31:26.120 --> 00:31:27.370 align:middle line:90% Let's get to that in a second. 00:31:27.370 --> 00:31:29.510 align:middle line:84% The first reason-- let's forget taxes. 00:31:29.510 --> 00:31:34.170 align:middle line:84% The first reason is because of the benefits of pooling. 00:31:34.170 --> 00:31:36.490 align:middle line:84% Remember insurance and adverse selection. 00:31:36.490 --> 00:31:39.350 align:middle line:84% Remember an insurer who's worried about adverse selection. 00:31:39.350 --> 00:31:40.810 align:middle line:90% What does that insurer want? 00:31:40.810 --> 00:31:42.250 align:middle line:90% To minimize the cost of-- 00:31:42.250 --> 00:31:45.770 align:middle line:84% to minimize the risk of adverse selection, they want two things. 00:31:45.770 --> 00:31:48.090 align:middle line:84% They want a big pool, and they want 00:31:48.090 --> 00:31:52.090 align:middle line:84% a pool that's constructed independent of health status. 00:31:52.090 --> 00:31:57.330 align:middle line:84% So you don't want the skydiving club 00:31:57.330 --> 00:31:59.223 align:middle line:90% But MIT, that's pretty good. 00:31:59.223 --> 00:32:00.640 align:middle line:84% People don't work at MIT-- they're 00:32:00.640 --> 00:32:01.990 align:middle line:84% not working at MIT because they're sick or healthy. 00:32:01.990 --> 00:32:03.790 align:middle line:84% We got some people, [INAUDIBLE] get some healthy people. 00:32:03.790 --> 00:32:04.590 align:middle line:90% That's fine. 00:32:04.590 --> 00:32:06.770 align:middle line:84% They want to live off the law of large numbers. 00:32:06.770 --> 00:32:08.370 align:middle line:84% That's basically what insurers want 00:32:08.370 --> 00:32:10.160 align:middle line:84% to do is they want to live off the law of large numbers. 00:32:10.160 --> 00:32:11.140 align:middle line:84% They want a big enough group they 00:32:11.140 --> 00:32:12.660 align:middle line:84% can rely on the law of large numbers 00:32:12.660 --> 00:32:14.860 align:middle line:84% and a symmetric enough group that they 00:32:14.860 --> 00:32:16.060 align:middle line:90% can map the distribution. 00:32:16.060 --> 00:32:19.380 align:middle line:84% With MIT, they can pretty much guess what our costs are 00:32:19.380 --> 00:32:20.675 align:middle line:90% going to be year after year. 00:32:20.675 --> 00:32:22.300 align:middle line:84% They can't predict the cancer, but they 00:32:22.300 --> 00:32:23.820 align:middle line:84% can know that X percent of our employees 00:32:23.820 --> 00:32:25.480 align:middle line:84% got cancer every year, statistically, 00:32:25.480 --> 00:32:29.157 align:middle line:84% and some year might be X plus 3% somebody X minus 3%. 00:32:29.157 --> 00:32:30.740 align:middle line:84% But roughly speaking, our costs aren't 00:32:30.740 --> 00:32:32.660 align:middle line:90% going to double next year. 00:32:32.660 --> 00:32:35.440 align:middle line:84% So insurers want large, predictable risks. 00:32:35.440 --> 00:32:37.740 align:middle line:84% And employers provide exactly that, especially 00:32:37.740 --> 00:32:38.840 align:middle line:90% large employers. 00:32:38.840 --> 00:32:42.313 align:middle line:90% 00:32:42.313 --> 00:32:47.100 align:middle line:84% As a result, large employers-- moreover, on top of that, 00:32:47.100 --> 00:32:49.020 align:middle line:84% the administrative costs of insurance 00:32:49.020 --> 00:32:52.140 align:middle line:90% have a large fixed component. 00:32:52.140 --> 00:32:54.060 align:middle line:84% So for both of those reasons, the bigger 00:32:54.060 --> 00:32:57.060 align:middle line:84% the group, the cheaper insurance can be, 00:32:57.060 --> 00:32:59.383 align:middle line:84% both because insurers can predict it better. 00:32:59.383 --> 00:33:00.800 align:middle line:84% So they have to charge less of a-- 00:33:00.800 --> 00:33:03.460 align:middle line:84% they can be more risk neutral, so they can charge a low price. 00:33:03.460 --> 00:33:06.300 align:middle line:84% And because administrative costs are lower per person. 00:33:06.300 --> 00:33:07.370 align:middle line:90% For both those reasons. 00:33:07.370 --> 00:33:10.630 align:middle line:84% Insurance is much cheaper for large firms than small firms. 00:33:10.630 --> 00:33:13.590 align:middle line:84% And for that reason, virtually every large firm in America 00:33:13.590 --> 00:33:14.630 align:middle line:90% offers health insurance. 00:33:14.630 --> 00:33:20.310 align:middle line:84% The fact is that 94% of firms with more than 50 employees, not 00:33:20.310 --> 00:33:23.090 align:middle line:84% even that big, even medium sized, offer health insurance. 00:33:23.090 --> 00:33:29.890 align:middle line:84% 94% with 50 employees and 99% with more than 1,000 employees. 00:33:29.890 --> 00:33:34.450 align:middle line:84% On the other hand, among firms with less than 50 employees, 00:33:34.450 --> 00:33:36.737 align:middle line:84% only 50% offer health insurance, and it shrinks. 00:33:36.737 --> 00:33:39.070 align:middle line:84% If you're down at firms that are less than 10 employees, 00:33:39.070 --> 00:33:41.830 align:middle line:84% only about 35% offer health insurance. 00:33:41.830 --> 00:33:44.130 align:middle line:84% And that's largely because you lose the risk pooling 00:33:44.130 --> 00:33:45.468 align:middle line:90% if you get too small. 00:33:45.468 --> 00:33:47.260 align:middle line:84% Even if your firm isn't the skydiving club, 00:33:47.260 --> 00:33:49.290 align:middle line:84% if it's only nine people and one gets cancer, 00:33:49.290 --> 00:33:50.990 align:middle line:90% that can double your costs. 00:33:50.990 --> 00:33:53.670 align:middle line:84% And so insurers are more wary of that. 00:33:53.670 --> 00:33:55.737 align:middle line:84% Insurers become more risk averse, if you will. 00:33:55.737 --> 00:33:58.070 align:middle line:84% They say, look, I'm going to charge you more because I'm 00:33:58.070 --> 00:33:59.320 align:middle line:90% worried about what's going on. 00:33:59.320 --> 00:34:01.350 align:middle line:84% I'm worried about adverse selection. 00:34:01.350 --> 00:34:07.040 align:middle line:84% So that's the first reason why employers are 00:34:07.040 --> 00:34:08.719 align:middle line:90% a source of health insurance. 00:34:08.719 --> 00:34:10.679 align:middle line:84% The second reason as well as mentioned, 00:34:10.679 --> 00:34:13.960 align:middle line:84% which is the tax subsidy employer health insurance, which 00:34:13.960 --> 00:34:20.040 align:middle line:84% is what MIT pays me in wages, I pay a tax rate 00:34:20.040 --> 00:34:23.320 align:middle line:84% that after my federal income tax plus Social Security plus state 00:34:23.320 --> 00:34:26.120 align:middle line:90% income tax adds up to about 50%. 00:34:26.120 --> 00:34:30.560 align:middle line:84% When MIT pays me in health insurance, I'm not taxed. 00:34:30.560 --> 00:34:33.560 align:middle line:90% It is a non-taxable benefit. 00:34:33.560 --> 00:34:36.620 align:middle line:84% As a result, if I'm going to get health insurance, 00:34:36.620 --> 00:34:38.360 align:middle line:84% I really want to get it through MIT. 00:34:38.360 --> 00:34:40.520 align:middle line:90% So to see that go to table 15-2. 00:34:40.520 --> 00:34:43.880 align:middle line:84% We have two individuals, Nigel and Khadija. 00:34:43.880 --> 00:34:47.540 align:middle line:84% They both have the same marginal product, $30,000, 00:34:47.540 --> 00:34:48.420 align:middle line:90% which is their wage. 00:34:48.420 --> 00:34:50.199 align:middle line:90% So it's a perfect labor market. 00:34:50.199 --> 00:34:52.199 align:middle line:90% They're paid their wage. 00:34:52.199 --> 00:34:55.060 align:middle line:84% But Nigel's employer doesn't provide health insurance. 00:34:55.060 --> 00:34:56.400 align:middle line:90% Khadija's does. 00:34:56.400 --> 00:34:57.817 align:middle line:84% The health insurance costs $5,000. 00:34:57.817 --> 00:35:00.192 align:middle line:84% So the employer, since it's a perfectly competitive labor 00:35:00.192 --> 00:35:02.520 align:middle line:84% market, just takes it out of Khadija's wages. 00:35:02.520 --> 00:35:06.130 align:middle line:84% So the employer says, we're going to pay Khadija $25,000 00:35:06.130 --> 00:35:09.970 align:middle line:84% in wage and $5,000 in health insurance. 00:35:09.970 --> 00:35:12.610 align:middle line:84% Now imagine there's a 33% tax rate. 00:35:12.610 --> 00:35:15.890 align:middle line:84% So after tax, Nigel takes home $20,000, 00:35:15.890 --> 00:35:19.850 align:middle line:90% and Khadija takes home 16,666. 00:35:19.850 --> 00:35:21.570 align:middle line:84% Now imagine insurance is actually 00:35:21.570 --> 00:35:24.910 align:middle line:84% cheaper outside the employer setting, which isn't true, 00:35:24.910 --> 00:35:26.450 align:middle line:90% but imagine it was. 00:35:26.450 --> 00:35:29.130 align:middle line:84% Imagine Nigel could actually get insurance outside employer 00:35:29.130 --> 00:35:31.367 align:middle line:90% setting for $4,000. 00:35:31.367 --> 00:35:32.950 align:middle line:84% So they say, great, he should do that. 00:35:32.950 --> 00:35:34.783 align:middle line:84% But look what happens at the end of the day, 00:35:34.783 --> 00:35:36.970 align:middle line:84% Nigel still ends up with less money. 00:35:36.970 --> 00:35:37.930 align:middle line:90% Why? 00:35:37.930 --> 00:35:41.970 align:middle line:84% Because he bought his insurance after paying taxes, 00:35:41.970 --> 00:35:43.770 align:middle line:84% whereas Khadija got to buy health insurance 00:35:43.770 --> 00:35:46.150 align:middle line:84% before paying taxes, so it became tax deductible. 00:35:46.150 --> 00:35:48.330 align:middle line:90% She wasn't taxed on it. 00:35:48.330 --> 00:35:50.650 align:middle line:84% Therefore, there is a bias towards getting your health 00:35:50.650 --> 00:35:53.090 align:middle line:90% insurance through your employer. 00:35:53.090 --> 00:35:54.670 align:middle line:84% Because if you get health insurance, 00:35:54.670 --> 00:35:58.690 align:middle line:84% you might as well take advantage of this tax break. 00:35:58.690 --> 00:36:03.750 align:middle line:84% Now here's a very important point, very important point 00:36:03.750 --> 00:36:04.890 align:middle line:90% that people misunderstand. 00:36:04.890 --> 00:36:06.910 align:middle line:84% If you understand it, you'll be in a very small, elite group 00:36:06.910 --> 00:36:08.110 align:middle line:90% of Americans who understand it. 00:36:08.110 --> 00:36:10.350 align:middle line:84% Most people don't know about the tax subsidy, but people who do 00:36:10.350 --> 00:36:11.210 align:middle line:90% get it wrong. 00:36:11.210 --> 00:36:14.350 align:middle line:84% This is not a tax break to employers. 00:36:14.350 --> 00:36:15.730 align:middle line:90% Employers don't care. 00:36:15.730 --> 00:36:18.350 align:middle line:84% It's a tax break to you, to the employee. 00:36:18.350 --> 00:36:20.830 align:middle line:84% So it's not that employers are treated differently 00:36:20.830 --> 00:36:21.430 align:middle line:90% if they give-- 00:36:21.430 --> 00:36:23.230 align:middle line:84% if I pay you wages or pay health insurance, either way, 00:36:23.230 --> 00:36:24.330 align:middle line:90% I'm paying you $30,000. 00:36:24.330 --> 00:36:25.430 align:middle line:90% I don't care. 00:36:25.430 --> 00:36:27.930 align:middle line:84% But as an employee, if I get paid wages or health insurance, 00:36:27.930 --> 00:36:28.450 align:middle line:90% I care. 00:36:28.450 --> 00:36:31.150 align:middle line:84% This is the employee tax subsidy. 00:36:31.150 --> 00:36:34.750 align:middle line:84% It's the employee tax subsidy to employer insurance, not 00:36:34.750 --> 00:36:36.263 align:middle line:90% the employer tax subsidy. 00:36:36.263 --> 00:36:38.430 align:middle line:84% It's a subsidy to individuals because the employer's 00:36:38.430 --> 00:36:38.930 align:middle line:90% indifferent. 00:36:38.930 --> 00:36:40.388 align:middle line:84% Either way, he's paying you 30,000. 00:36:40.388 --> 00:36:41.350 align:middle line:90% What does he care? 00:36:41.350 --> 00:36:46.630 align:middle line:84% So that's the major way people get their health insurance. 00:36:46.630 --> 00:36:48.890 align:middle line:84% The second major way is the nongroup market. 00:36:48.890 --> 00:36:51.270 align:middle line:84% As I said before, the Affordable Care Act, 00:36:51.270 --> 00:36:53.893 align:middle line:90% this is a very messed up market. 00:36:53.893 --> 00:36:55.310 align:middle line:84% And I'll talk more about that when 00:36:55.310 --> 00:36:56.893 align:middle line:84% we talk about the Affordable Care Act, 00:36:56.893 --> 00:36:59.470 align:middle line:90% probably in two lectures. 00:36:59.470 --> 00:37:03.560 align:middle line:84% But now, it's a fairly successful market. 00:37:03.560 --> 00:37:05.840 align:middle line:84% We then have the two big government programs. 00:37:05.840 --> 00:37:06.620 align:middle line:90% Yeah? 00:37:06.620 --> 00:37:11.620 align:middle line:84% AUDIENCE: The [INAUDIBLE] has this ripple effect on wages. 00:37:11.620 --> 00:37:14.080 align:middle line:84% Where does that subsidy actually really [? land? ?] 00:37:14.080 --> 00:37:16.200 align:middle line:90% JON GRUBER: On the taxpayer. 00:37:16.200 --> 00:37:20.680 align:middle line:84% So if you think about this, that basically Khadija is getting 00:37:20.680 --> 00:37:26.120 align:middle line:84% the same compensation as Nigel, $30,000. 00:37:26.120 --> 00:37:29.480 align:middle line:84% But the US tax system is giving her a lot more money. 00:37:29.480 --> 00:37:33.600 align:middle line:84% So if you add up the loss-- if health insurance was taxed 00:37:33.600 --> 00:37:36.240 align:middle line:84% as wages in the US, we would raise 00:37:36.240 --> 00:37:41.400 align:middle line:84% almost $350 billion every single year more in tax revenues. 00:37:41.400 --> 00:37:42.760 align:middle line:90% That's a lot. 00:37:42.760 --> 00:37:45.840 align:middle line:84% So basically, this is a huge, huge tax break. 00:37:45.840 --> 00:37:49.120 align:middle line:90% 00:37:49.120 --> 00:37:52.200 align:middle line:84% Because there's no-- I'll come back to this in chapter 18. 00:37:52.200 --> 00:37:53.680 align:middle line:90% It's just compensation. 00:37:53.680 --> 00:37:56.180 align:middle line:84% We're just saying one form of compensation gets a tax break. 00:37:56.180 --> 00:37:56.580 align:middle line:90% One doesn't. 00:37:56.580 --> 00:37:58.622 align:middle line:84% If we treated all forms of compensation the same, 00:37:58.622 --> 00:37:59.950 align:middle line:90% we'd raise a lot more revenue. 00:37:59.950 --> 00:38:01.950 align:middle line:84% AUDIENCE: Wages being able to fluctuate, though, 00:38:01.950 --> 00:38:03.918 align:middle line:84% don't employers claim some of that, at least? 00:38:03.918 --> 00:38:05.710 align:middle line:84% JON GRUBER: Well, that's the open question. 00:38:05.710 --> 00:38:07.470 align:middle line:84% Does it all get shifted to wages or not? 00:38:07.470 --> 00:38:10.050 align:middle line:84% And the evidence looks like it all gets shifted to wages. 00:38:10.050 --> 00:38:10.850 align:middle line:90% Yeah? 00:38:10.850 --> 00:38:12.230 align:middle line:84% AUDIENCE: What's the rationale for doing this? 00:38:12.230 --> 00:38:13.690 align:middle line:90% Was it to induce people to buy-- 00:38:13.690 --> 00:38:18.083 align:middle line:84% JON GRUBER: The history of it is that during World War II, 00:38:18.083 --> 00:38:19.750 align:middle line:84% the government was worried about prices, 00:38:19.750 --> 00:38:22.950 align:middle line:84% so they put in place wage limits. 00:38:22.950 --> 00:38:24.890 align:middle line:84% You couldn't raise your employees' wages. 00:38:24.890 --> 00:38:27.330 align:middle line:84% So employers got around that by introducing benefits 00:38:27.330 --> 00:38:29.747 align:middle line:84% as a way to compensate their employees more. 00:38:29.747 --> 00:38:31.330 align:middle line:84% And then since those were new, the IRS 00:38:31.330 --> 00:38:33.090 align:middle line:84% said, well, I guess they're not taxable. 00:38:33.090 --> 00:38:34.790 align:middle line:84% Because it's some new thing, we're not going to tax it. 00:38:34.790 --> 00:38:35.915 align:middle line:90% And that became entrenched. 00:38:35.915 --> 00:38:37.430 align:middle line:90% So pensions are the same thing. 00:38:37.430 --> 00:38:40.990 align:middle line:84% Many other employee benefits are the same thing. 00:38:40.990 --> 00:38:41.490 align:middle line:90% OK. 00:38:41.490 --> 00:38:43.490 align:middle line:84% So then we have the two big government programs. 00:38:43.490 --> 00:38:44.850 align:middle line:90% We have Medicare and Medicaid. 00:38:44.850 --> 00:38:49.410 align:middle line:84% Now, look, I probably won't run you run into you later in life. 00:38:49.410 --> 00:38:51.770 align:middle line:84% But if I do, there's only one thing 00:38:51.770 --> 00:38:53.530 align:middle line:84% I'm going to hold you responsible for, 00:38:53.530 --> 00:38:56.700 align:middle line:84% which is knowing the difference between Medicare and Medicaid. 00:38:56.700 --> 00:39:00.420 align:middle line:84% Medicare is insurance for the elderly. 00:39:00.420 --> 00:39:03.660 align:middle line:84% Medicaid is insurance for the poor. 00:39:03.660 --> 00:39:06.580 align:middle line:84% Now, there are poor elderly, who get both. 00:39:06.580 --> 00:39:08.820 align:middle line:84% But the bottom line is, think of it that way. 00:39:08.820 --> 00:39:10.260 align:middle line:84% Medicare is insurance-- actually, 00:39:10.260 --> 00:39:11.468 align:middle line:90% for the elderly and disabled. 00:39:11.468 --> 00:39:13.220 align:middle line:84% I won't hold you responsible for that. 00:39:13.220 --> 00:39:17.640 align:middle line:84% But if you're over 65 or on disability insurance-- 00:39:17.640 --> 00:39:19.780 align:middle line:90% we talked about last class-- 00:39:19.780 --> 00:39:24.500 align:middle line:84% you qualify for Medicare, which is a single-payer government 00:39:24.500 --> 00:39:25.000 align:middle line:90% program. 00:39:25.000 --> 00:39:26.917 align:middle line:84% It's our version of what the rest of the world 00:39:26.917 --> 00:39:27.940 align:middle line:90% has for everybody. 00:39:27.940 --> 00:39:31.420 align:middle line:84% It's a single-payer government program. 00:39:31.420 --> 00:39:33.100 align:middle line:84% Or at least, it was traditionally. 00:39:33.100 --> 00:39:36.740 align:middle line:84% It's changing over time-- was traditionally. 00:39:36.740 --> 00:39:38.960 align:middle line:84% This program, it's paid for by payroll tax. 00:39:38.960 --> 00:39:41.720 align:middle line:84% You'll notice on your pay stub, in addition to the FICA row, 00:39:41.720 --> 00:39:45.143 align:middle line:84% there's a row for the Medicare tax, 1.45%. 00:39:45.143 --> 00:39:47.560 align:middle line:84% The difference with the Medicare tax, unlike the FICA tax, 00:39:47.560 --> 00:39:49.340 align:middle line:90% it's uncapped. 00:39:49.340 --> 00:39:54.620 align:middle line:84% You pay that 1.45% all the way on your income. 00:39:54.620 --> 00:39:57.950 align:middle line:84% Indeed, as part of the Affordable Care Act, 00:39:57.950 --> 00:40:00.450 align:middle line:84% they actually raised that by another 0.9% 00:40:00.450 --> 00:40:07.870 align:middle line:84% So if your income's above $250,000, it's actually 2.35%. 00:40:07.870 --> 00:40:11.350 align:middle line:84% So basically, the way it works is, 00:40:11.350 --> 00:40:16.430 align:middle line:84% if you've worked for 10 years in Medicare-covered employment-- 00:40:16.430 --> 00:40:17.890 align:middle line:90% so this is like Social Security. 00:40:17.890 --> 00:40:20.450 align:middle line:84% If you work long enough to qualify, then when you turn 65, 00:40:20.450 --> 00:40:21.750 align:middle line:90% you get Medicare. 00:40:21.750 --> 00:40:24.430 align:middle line:84% It's just an entitlement for turning 65. 00:40:24.430 --> 00:40:27.090 align:middle line:84% And I'm going to come back and talk a lot more about Medicare. 00:40:27.090 --> 00:40:29.007 align:middle line:84% I just want to give the basic lay of the land. 00:40:29.007 --> 00:40:30.590 align:middle line:84% That's Medicare, so for the elderly. 00:40:30.590 --> 00:40:33.550 align:middle line:84% Medicaid is much more complicated 00:40:33.550 --> 00:40:37.630 align:middle line:84% because Medicaid is targeted at several different groups. 00:40:37.630 --> 00:40:40.550 align:middle line:84% So the groups we'll focus on, and where most of the people 00:40:40.550 --> 00:40:45.470 align:middle line:84% are, are low-income children and families. 00:40:45.470 --> 00:40:47.910 align:middle line:90% That is traditionally. 00:40:47.910 --> 00:40:50.608 align:middle line:84% So traditionally, when Medicaid was set up, 00:40:50.608 --> 00:40:52.150 align:middle line:84% it was the health insurance component 00:40:52.150 --> 00:40:53.817 align:middle line:84% that went along with the cash welfare we 00:40:53.817 --> 00:40:55.560 align:middle line:90% delivered to very poor people. 00:40:55.560 --> 00:40:57.840 align:middle line:84% It was originally for just very poor people. 00:40:57.840 --> 00:41:04.020 align:middle line:84% In the 1980s and 1990s, it got expanded not just to poor people 00:41:04.020 --> 00:41:07.440 align:middle line:84% but to higher-income people for pregnancy and children. 00:41:07.440 --> 00:41:10.000 align:middle line:84% So children became eligible if their incomes were higher 00:41:10.000 --> 00:41:13.400 align:middle line:84% and women became eligible for the expenses of pregnancy. 00:41:13.400 --> 00:41:15.720 align:middle line:90% So it was very poor families. 00:41:15.720 --> 00:41:18.480 align:middle line:84% Then it was expanded more for up to middle class 00:41:18.480 --> 00:41:21.320 align:middle line:90% for kids and pregnant women. 00:41:21.320 --> 00:41:25.140 align:middle line:84% And then as part of the Affordable Care Act, 00:41:25.140 --> 00:41:28.745 align:middle line:84% we made everyone eligible up to 133% of the poverty line. 00:41:28.745 --> 00:41:30.620 align:middle line:84% We haven't talked about the poverty line yet, 00:41:30.620 --> 00:41:34.320 align:middle line:84% but 100% of poverty line is about $16,000 for a person 00:41:34.320 --> 00:41:36.640 align:middle line:90% or $32,000 for a family. 00:41:36.640 --> 00:41:40.520 align:middle line:84% So basically, Medicaid today is universal coverage 00:41:40.520 --> 00:41:44.960 align:middle line:84% for people who are poor or near poor in the US. 00:41:44.960 --> 00:41:47.580 align:middle line:90% That is most of the people. 00:41:47.580 --> 00:41:51.400 align:middle line:84% That is 3/4 of the people on Medicaid. 00:41:51.400 --> 00:41:54.730 align:middle line:84% It is only about 1/3 of the spending. 00:41:54.730 --> 00:41:56.530 align:middle line:84% Most of the spending is the last group 00:41:56.530 --> 00:41:59.290 align:middle line:84% covered by Medicaid, which is the people who 00:41:59.290 --> 00:42:03.250 align:middle line:84% are people who are using long-term care services, nursing 00:42:03.250 --> 00:42:06.047 align:middle line:84% homes and home care, mostly elderly. 00:42:06.047 --> 00:42:07.630 align:middle line:84% That's why I said there's the overlap. 00:42:07.630 --> 00:42:10.010 align:middle line:84% The poor elderly get both Medicare and Medicaid, 00:42:10.010 --> 00:42:11.190 align:middle line:90% but not exclusively elderly. 00:42:11.190 --> 00:42:12.470 align:middle line:90% They include disability. 00:42:12.470 --> 00:42:14.010 align:middle line:84% That includes some people who are 00:42:14.010 --> 00:42:17.150 align:middle line:84% disabled who, say, are mentally disabled, never worked at all. 00:42:17.150 --> 00:42:19.530 align:middle line:90% They'll also be on Medicaid. 00:42:19.530 --> 00:42:21.943 align:middle line:84% So essentially, most of Medicaid spending 00:42:21.943 --> 00:42:23.610 align:middle line:84% is actually on people who are in nursing 00:42:23.610 --> 00:42:25.490 align:middle line:84% homes and other institutions, as well as 00:42:25.490 --> 00:42:27.450 align:middle line:90% people getting home care. 00:42:27.450 --> 00:42:30.510 align:middle line:90% But most of the people-- 00:42:30.510 --> 00:42:32.050 align:middle line:84% it's sort of a universal entitlement 00:42:32.050 --> 00:42:34.530 align:middle line:90% to low-income Americans. 00:42:34.530 --> 00:42:39.330 align:middle line:90% So that's Medicaid. 00:42:39.330 --> 00:42:42.290 align:middle line:84% So if you look at your table, Medicare and Medicaid 00:42:42.290 --> 00:42:47.210 align:middle line:84% are each about 56, 59 million people on those programs. 00:42:47.210 --> 00:42:52.830 align:middle line:84% Then the last cartridge category of insurance is Tricare. 00:42:52.830 --> 00:42:54.830 align:middle line:84% Tricare is insurance for those in the military 00:42:54.830 --> 00:42:56.630 align:middle line:90% and their dependents. 00:42:56.630 --> 00:42:58.510 align:middle line:90% It's a small number of people. 00:42:58.510 --> 00:43:01.070 align:middle line:84% And then, finally, we have the uninsured. 00:43:01.070 --> 00:43:02.910 align:middle line:90% So who are the uninsured? 00:43:02.910 --> 00:43:06.670 align:middle line:84% Well, the uninsured are poor, but they're typically not 00:43:06.670 --> 00:43:10.990 align:middle line:84% the poorest because the poorest can get free Medicaid. 00:43:10.990 --> 00:43:13.670 align:middle line:84% You could think of a typical uninsured person 00:43:13.670 --> 00:43:16.510 align:middle line:90% as falling into two groups. 00:43:16.510 --> 00:43:19.230 align:middle line:84% One is a very poor person who simply 00:43:19.230 --> 00:43:21.310 align:middle line:90% hasn't taken up their Medicaid. 00:43:21.310 --> 00:43:23.310 align:middle line:84% Remember, I talked about the problem of take-up. 00:43:23.310 --> 00:43:26.790 align:middle line:84% Some people don't take in things they're entitled to. 00:43:26.790 --> 00:43:29.097 align:middle line:84% Probably the majority of the uninsured in America, 00:43:29.097 --> 00:43:30.930 align:middle line:84% or a large share of the uninsured, maybe not 00:43:30.930 --> 00:43:33.270 align:middle line:84% the majority, but near majority, could get free health insurance 00:43:33.270 --> 00:43:34.020 align:middle line:90% if they wanted to. 00:43:34.020 --> 00:43:35.310 align:middle line:90% They just don't take it up. 00:43:35.310 --> 00:43:36.750 align:middle line:90% So that's one group. 00:43:36.750 --> 00:43:40.630 align:middle line:84% The other group is what we might call the working poor. 00:43:40.630 --> 00:43:44.970 align:middle line:84% These are people who work, earn more than 133% of poverty, 00:43:44.970 --> 00:43:47.630 align:middle line:84% so can't get Medicaid, but don't work for employers that 00:43:47.630 --> 00:43:50.088 align:middle line:84% offer health insurance or can't afford the health insurance 00:43:50.088 --> 00:43:51.160 align:middle line:90% their employer offers. 00:43:51.160 --> 00:43:57.260 align:middle line:84% That's a big chunk of the uninsured, is that group. 00:43:57.260 --> 00:43:57.760 align:middle line:90% OK. 00:43:57.760 --> 00:44:02.600 align:middle line:90% So why are people uninsured? 00:44:02.600 --> 00:44:05.820 align:middle line:84% Well, remember, in chapter 12, we said, 00:44:05.820 --> 00:44:08.000 align:middle line:84% look, the optimum is full insurance. 00:44:08.000 --> 00:44:09.685 align:middle line:84% People should want full insurance. 00:44:09.685 --> 00:44:11.060 align:middle line:84% Why aren't they buying insurance? 00:44:11.060 --> 00:44:13.200 align:middle line:84% Well, there's a number of reasons. 00:44:13.200 --> 00:44:15.240 align:middle line:84% The first reason could be that insurance is not 00:44:15.240 --> 00:44:16.640 align:middle line:90% actuarially fair. 00:44:16.640 --> 00:44:20.200 align:middle line:84% Remember, we said full insurance was optimal 00:44:20.200 --> 00:44:24.960 align:middle line:84% under a situation of zero-profit insurance being sold, 00:44:24.960 --> 00:44:27.360 align:middle line:84% where insurers sold insurance at exactly what it 00:44:27.360 --> 00:44:28.920 align:middle line:90% cost them to deliver the care. 00:44:28.920 --> 00:44:29.660 align:middle line:90% That's not true. 00:44:29.660 --> 00:44:30.600 align:middle line:84% As I said a few minutes ago, we have 00:44:30.600 --> 00:44:33.200 align:middle line:84% about an average about a 15% administrative load on insurance 00:44:33.200 --> 00:44:34.280 align:middle line:90% in America. 00:44:34.280 --> 00:44:36.360 align:middle line:84% Therefore, many people don't find it worth it 00:44:36.360 --> 00:44:39.720 align:middle line:84% because it's 15% more than what will be actuarially fair. 00:44:39.720 --> 00:44:41.840 align:middle line:90% So that's one reason. 00:44:41.840 --> 00:44:45.560 align:middle line:84% The second reason is adverse selection. 00:44:45.560 --> 00:44:48.130 align:middle line:84% Remember, we talked about adverse selection 00:44:48.130 --> 00:44:49.730 align:middle line:90% in health insurance markets. 00:44:49.730 --> 00:44:54.367 align:middle line:84% And we talked about the fact that because only 00:44:54.367 --> 00:44:55.950 align:middle line:84% the sick people want to buy insurance, 00:44:55.950 --> 00:44:58.117 align:middle line:84% and insurers know that, they may not offer insurance 00:44:58.117 --> 00:45:00.130 align:middle line:84% or may not offer complete insurance. 00:45:00.130 --> 00:45:04.050 align:middle line:84% That's the second reason they'll be uninsured people. 00:45:04.050 --> 00:45:08.010 align:middle line:84% A third reason is, it could be a completely rational decision 00:45:08.010 --> 00:45:11.010 align:middle line:90% to forgo insurance. 00:45:11.010 --> 00:45:15.010 align:middle line:84% Once again, insurance is expensive. 00:45:15.010 --> 00:45:17.110 align:middle line:84% If you're a 25-year-old male, and once again, 00:45:17.110 --> 00:45:19.360 align:middle line:84% I'm saying male because there's no risk of pregnancy-- 00:45:19.360 --> 00:45:21.210 align:middle line:90% 25-year-old single male. 00:45:21.210 --> 00:45:23.850 align:middle line:84% Basically, the main risk is getting 00:45:23.850 --> 00:45:26.730 align:middle line:84% hit by a car or some other catastrophe. 00:45:26.730 --> 00:45:29.450 align:middle line:84% And if that happens, you can go to the hospital, 00:45:29.450 --> 00:45:31.570 align:middle line:84% and the hospital has to treat you. 00:45:31.570 --> 00:45:34.690 align:middle line:84% We have a law in America which says that if someone shows up 00:45:34.690 --> 00:45:36.990 align:middle line:84% at the emergency room, you have to treat them, 00:45:36.990 --> 00:45:39.602 align:middle line:84% regardless of their insurance status. 00:45:39.602 --> 00:45:41.310 align:middle line:84% So for many people, they might say, look, 00:45:41.310 --> 00:45:43.930 align:middle line:84% if my only risk is something send me in the emergency room, 00:45:43.930 --> 00:45:45.430 align:middle line:84% then why should I pay for insurance? 00:45:45.430 --> 00:45:48.340 align:middle line:84% If I have to go, the emergency room will be free anyway. 00:45:48.340 --> 00:45:52.660 align:middle line:84% So basically, there's what we might call rational free riding. 00:45:52.660 --> 00:45:54.380 align:middle line:84% Because when people go to emergency room, 00:45:54.380 --> 00:45:55.338 align:middle line:90% it may be free to them. 00:45:55.338 --> 00:45:57.220 align:middle line:90% It's not free to society. 00:45:57.220 --> 00:46:00.780 align:middle line:84% That becomes what we call uncompensated care, 00:46:00.780 --> 00:46:08.940 align:middle line:84% uncompensated care, which is basically care 00:46:08.940 --> 00:46:11.420 align:middle line:84% that hospitals deliver they're not reimbursed for. 00:46:11.420 --> 00:46:14.500 align:middle line:84% And that is an externality on the rest of us 00:46:14.500 --> 00:46:17.140 align:middle line:84% because someone has to pay those costs. 00:46:17.140 --> 00:46:17.940 align:middle line:90% Yeah? 00:46:17.940 --> 00:46:19.998 align:middle line:84% AUDIENCE: And it's free to them because they 00:46:19.998 --> 00:46:21.540 align:middle line:84% don't expect the hospital to sue them 00:46:21.540 --> 00:46:23.240 align:middle line:84% or because they could go bankrupt. 00:46:23.240 --> 00:46:24.235 align:middle line:90% And then [INAUDIBLE]. 00:46:24.235 --> 00:46:24.860 align:middle line:90% JON GRUBER: No. 00:46:24.860 --> 00:46:27.027 align:middle line:84% The hospital's bill them, but people just don't pay. 00:46:27.027 --> 00:46:29.358 align:middle line:84% Turns out, it's pretty hard to get people to pay. 00:46:29.358 --> 00:46:31.900 align:middle line:84% There's a paper recently about someone who went to a hospital 00:46:31.900 --> 00:46:34.700 align:middle line:84% and tried to buy the debt that people owed. 00:46:34.700 --> 00:46:37.060 align:middle line:84% And the hospital will sell it to them at 5 and 1/2 cents 00:46:37.060 --> 00:46:38.860 align:middle line:90% on the dollar. 00:46:38.860 --> 00:46:40.820 align:middle line:84% So basically, most people just don't pay. 00:46:40.820 --> 00:46:42.380 align:middle line:84% AUDIENCE: So they'll make it that that individual 00:46:42.380 --> 00:46:44.213 align:middle line:84% and see it as like [INAUDIBLE] credit score? 00:46:44.213 --> 00:46:47.510 align:middle line:84% JON GRUBER: Yeah, the credit score or something like that. 00:46:47.510 --> 00:46:51.652 align:middle line:90% So that's another reason. 00:46:51.652 --> 00:46:53.610 align:middle line:84% Another reason is, people just can't afford it. 00:46:53.610 --> 00:46:55.290 align:middle line:84% There's a [INAUDIBLE] liquidity constraints. 00:46:55.290 --> 00:46:57.082 align:middle line:84% And you can't borrow on your human capital. 00:46:57.082 --> 00:47:00.170 align:middle line:84% So maybe health insurance would be an optimal investment. 00:47:00.170 --> 00:47:01.950 align:middle line:90% They can't afford it. 00:47:01.950 --> 00:47:03.970 align:middle line:84% And then there's just standard paternalism. 00:47:03.970 --> 00:47:05.885 align:middle line:84% People may just be making bad decisions. 00:47:05.885 --> 00:47:07.510 align:middle line:84% Maybe they should have health insurance 00:47:07.510 --> 00:47:09.230 align:middle line:84% and they're just not appropriately calculating 00:47:09.230 --> 00:47:10.730 align:middle line:84% the risk and understanding the risks 00:47:10.730 --> 00:47:12.930 align:middle line:90% they face if they're uninsured. 00:47:12.930 --> 00:47:13.430 align:middle line:90% Yeah? 00:47:13.430 --> 00:47:16.710 align:middle line:84% AUDIENCE: What about very healthy, very rich people 00:47:16.710 --> 00:47:20.490 align:middle line:84% that see themselves not needing health care frequently at all? 00:47:20.490 --> 00:47:23.518 align:middle line:84% But in the case where they do, they could dish out the money? 00:47:23.518 --> 00:47:24.810 align:middle line:90% JON GRUBER: It's a great point. 00:47:24.810 --> 00:47:28.670 align:middle line:84% Remember, insurance loses its value 00:47:28.670 --> 00:47:31.950 align:middle line:84% as the risk becomes small relative to your resources. 00:47:31.950 --> 00:47:36.670 align:middle line:84% So if you are a billionaire and the largest possible expense 00:47:36.670 --> 00:47:39.950 align:middle line:84% that could ever happen is $100,000, 00:47:39.950 --> 00:47:41.420 align:middle line:84% then you're roughly risk neutral. 00:47:41.420 --> 00:47:43.170 align:middle line:84% So why pay money to the insurance company? 00:47:43.170 --> 00:47:44.520 align:middle line:90% Why don't you pay out of pocket? 00:47:44.520 --> 00:47:48.100 align:middle line:84% The problem is, the largest expense is not $100,000. 00:47:48.100 --> 00:47:50.380 align:middle line:84% You can have a medical bill of $10 million. 00:47:50.380 --> 00:47:52.480 align:middle line:84% Even for a billionaire, that's real money. 00:47:52.480 --> 00:47:55.560 align:middle line:84% And we're going to come back to what kind of insurance that 00:47:55.560 --> 00:47:56.660 align:middle line:90% suggests they should buy. 00:47:56.660 --> 00:47:58.300 align:middle line:84% You can probably figure that out already. 00:47:58.300 --> 00:48:00.160 align:middle line:84% But that has features of what kind of insurance 00:48:00.160 --> 00:48:00.860 align:middle line:90% they should buy. 00:48:00.860 --> 00:48:05.240 align:middle line:90% 00:48:05.240 --> 00:48:06.700 align:middle line:84% So that's why people are uninsured. 00:48:06.700 --> 00:48:09.783 align:middle line:84% What's more interesting is, why do we care? 00:48:09.783 --> 00:48:12.200 align:middle line:84% Like I said in the beginning of the course, lots of people 00:48:12.200 --> 00:48:14.800 align:middle line:90% don't have stuff in America. 00:48:14.800 --> 00:48:17.140 align:middle line:84% We don't feel like we have to fix all that stuff. 00:48:17.140 --> 00:48:20.080 align:middle line:84% Why do we feel we have to fix the uninsurance problem? 00:48:20.080 --> 00:48:23.280 align:middle line:84% And one reason I gave-- so why do we care? 00:48:23.280 --> 00:48:25.420 align:middle line:90% Why care? 00:48:25.420 --> 00:48:29.550 align:middle line:84% One reason I gave the first lecture was externalities, 00:48:29.550 --> 00:48:31.300 align:middle line:84% which is that basically, people uninsured, 00:48:31.300 --> 00:48:32.680 align:middle line:90% they might get you sick. 00:48:32.680 --> 00:48:35.060 align:middle line:84% Or alternatively, they might get uncompensated care, 00:48:35.060 --> 00:48:38.043 align:middle line:84% which amounts to about $30 billion a year. 00:48:38.043 --> 00:48:39.960 align:middle line:84% And typically, our view is that the rest of us 00:48:39.960 --> 00:48:43.140 align:middle line:84% pay that $30 billion through taxes or through higher 00:48:43.140 --> 00:48:45.180 align:middle line:90% insurance prices or whatever. 00:48:45.180 --> 00:48:54.660 align:middle line:84% So the first reason might care is externalities, so basically, 00:48:54.660 --> 00:48:59.460 align:middle line:84% either physical or financial externalities. 00:48:59.460 --> 00:49:06.620 align:middle line:84% The second reason is that by people being uninsured, 00:49:06.620 --> 00:49:08.940 align:middle line:90% they don't get care efficiently. 00:49:08.940 --> 00:49:14.420 align:middle line:84% And ultimately, in some sense-- externality is sort of 1a is 00:49:14.420 --> 00:49:17.940 align:middle line:84% inefficiencies, which is, if people are uninsured, 00:49:17.940 --> 00:49:20.940 align:middle line:84% they then get sick, and then that imposes big externalities 00:49:20.940 --> 00:49:22.020 align:middle line:90% once they're sick. 00:49:22.020 --> 00:49:24.660 align:middle line:84% So for example, there was a famous study 00:49:24.660 --> 00:49:26.910 align:middle line:84% I loved back from when I was in grad school, where 00:49:26.910 --> 00:49:28.660 align:middle line:84% a guy went to an emergency room and talked 00:49:28.660 --> 00:49:29.640 align:middle line:90% to people sitting around. 00:49:29.640 --> 00:49:31.160 align:middle line:84% Obviously, these aren't the worst cases. 00:49:31.160 --> 00:49:32.120 align:middle line:90% They're sitting around. 00:49:32.120 --> 00:49:34.760 align:middle line:84% But most people in the emergency room are sitting around. 00:49:34.760 --> 00:49:36.580 align:middle line:84% And he said to them, how many of you 00:49:36.580 --> 00:49:40.310 align:middle line:84% would happily trade this visit for a doctor's appointment 00:49:40.310 --> 00:49:42.030 align:middle line:90% in three days? 00:49:42.030 --> 00:49:45.772 align:middle line:90% And like 70% said yes. 00:49:45.772 --> 00:49:47.230 align:middle line:84% Well, what's the actual percentage? 00:49:47.230 --> 00:49:48.230 align:middle line:90% I have that here. 00:49:48.230 --> 00:49:51.492 align:middle line:84% The actual-- I'm sorry, 40% said yes. 00:49:51.492 --> 00:49:52.950 align:middle line:84% Because basically, they didn't need 00:49:52.950 --> 00:49:54.390 align:middle line:90% to go to the emergency room. 00:49:54.390 --> 00:49:55.370 align:middle line:90% They just didn't have insurance. 00:49:55.370 --> 00:49:57.995 align:middle line:84% And doctors don't have to take you if you don't have insurance. 00:49:57.995 --> 00:49:59.190 align:middle line:90% Emergency rooms do. 00:49:59.190 --> 00:50:00.107 align:middle line:90% So that's inefficient. 00:50:00.107 --> 00:50:01.940 align:middle line:84% They should not be getting their doctor care 00:50:01.940 --> 00:50:02.950 align:middle line:90% in the emergency room. 00:50:02.950 --> 00:50:05.910 align:middle line:90% But why not for them? 00:50:05.910 --> 00:50:07.550 align:middle line:84% So basically, there's inefficiency 00:50:07.550 --> 00:50:09.925 align:middle line:84% that arises from people being uninsured, which ultimately 00:50:09.925 --> 00:50:12.190 align:middle line:84% leads to these externalities through health 00:50:12.190 --> 00:50:14.790 align:middle line:90% or through fiscal externalities. 00:50:14.790 --> 00:50:19.630 align:middle line:84% Then there's the reason that we might actually 00:50:19.630 --> 00:50:23.710 align:middle line:90% feel altruism, that basically. 00:50:23.710 --> 00:50:27.390 align:middle line:84% We feel that people are uninsured are suffering. 00:50:27.390 --> 00:50:28.730 align:middle line:90% Maybe we're paternalistic. 00:50:28.730 --> 00:50:30.210 align:middle line:84% We feel they're not making the right decision. 00:50:30.210 --> 00:50:31.335 align:middle line:90% And we feel badly for them. 00:50:31.335 --> 00:50:33.830 align:middle line:84% We want to make sure they have health insurance. 00:50:33.830 --> 00:50:37.190 align:middle line:84% So that's another reason we might care. 00:50:37.190 --> 00:50:41.120 align:middle line:84% But maybe be economically-- so there's no altruism slash 00:50:41.120 --> 00:50:42.340 align:middle line:90% paternalism. 00:50:42.340 --> 00:50:46.000 align:middle line:90% 00:50:46.000 --> 00:50:48.440 align:middle line:84% But there's another economic reason we might care, 00:50:48.440 --> 00:50:54.240 align:middle line:84% which is unique American problem of job lock, which 00:50:54.240 --> 00:50:56.160 align:middle line:84% is that since people get their health 00:50:56.160 --> 00:50:58.200 align:middle line:84% insurance through their job, people 00:50:58.200 --> 00:51:01.040 align:middle line:84% may be unwilling to leave their job if they 00:51:01.040 --> 00:51:03.108 align:middle line:90% risk becoming uninsured. 00:51:03.108 --> 00:51:05.400 align:middle line:84% So I like to write down a simple model that illustrates 00:51:05.400 --> 00:51:07.040 align:middle line:90% this that makes it very clear. 00:51:07.040 --> 00:51:11.900 align:middle line:84% So imagine that Emma has utility over two goods, consumption, 00:51:11.900 --> 00:51:12.775 align:middle line:90% health insurance. 00:51:12.775 --> 00:51:13.900 align:middle line:90% She likes health insurance. 00:51:13.900 --> 00:51:15.192 align:middle line:90% Just think of health insurance. 00:51:15.192 --> 00:51:17.460 align:middle line:84% She's really valuing not having to pay her bills, 00:51:17.460 --> 00:51:18.520 align:middle line:84% but just making it easier to say there's 00:51:18.520 --> 00:51:20.100 align:middle line:90% two goods of utility function. 00:51:20.100 --> 00:51:25.560 align:middle line:84% So it's U of C comma HI, where HI is 1 or 0, whether or not 00:51:25.560 --> 00:51:27.240 align:middle line:90% she has health insurance. 00:51:27.240 --> 00:51:32.760 align:middle line:84% Suppose we're in a perfectly well-functioning labor market 00:51:32.760 --> 00:51:35.760 align:middle line:84% where compensation equals your marginal product. 00:51:35.760 --> 00:51:38.355 align:middle line:84% And as I said, in such a market, if your employer gives you 00:51:38.355 --> 00:51:40.730 align:middle line:84% health insurance, they'll just take it out of your wages. 00:51:40.730 --> 00:51:42.610 align:middle line:84% That's the way a perfectly functioning labor 00:51:42.610 --> 00:51:45.050 align:middle line:90% market would work. 00:51:45.050 --> 00:51:48.010 align:middle line:84% So suppose that Emma currently works 00:51:48.010 --> 00:51:50.310 align:middle line:84% for a-- what are they, Big Three now, accounting firms? 00:51:50.310 --> 00:51:50.930 align:middle line:90% They used to be Big Seven. 00:51:50.930 --> 00:51:52.263 align:middle line:90% I don't know what they are now-- 00:51:52.263 --> 00:51:53.610 align:middle line:90% Big Four accounting firm. 00:51:53.610 --> 00:51:55.150 align:middle line:90% It's a good job. 00:51:55.150 --> 00:51:56.730 align:middle line:90% She gets health insurance. 00:51:56.730 --> 00:52:00.410 align:middle line:84% Now suppose she has a great idea for a startup, where she could 00:52:00.410 --> 00:52:02.530 align:middle line:84% go and make a shitload more money, 00:52:02.530 --> 00:52:05.210 align:middle line:84% but she'd lose her health insurance. 00:52:05.210 --> 00:52:06.690 align:middle line:90% So what's her decision? 00:52:06.690 --> 00:52:11.850 align:middle line:84% Well, her decision is, do I work in job 1, 00:52:11.850 --> 00:52:13.350 align:middle line:90% with a marginal product of 1? 00:52:13.350 --> 00:52:15.330 align:middle line:84% So our choice is utility of working 00:52:15.330 --> 00:52:20.690 align:middle line:84% in job 1, where she has a marginal product of 1, MP1. 00:52:20.690 --> 00:52:22.990 align:middle line:84% She pays the cost of health insurance. 00:52:22.990 --> 00:52:25.650 align:middle line:90% Let's call that C. 00:52:25.650 --> 00:52:29.050 align:middle line:90% But she gets health insurance-- 00:52:29.050 --> 00:52:33.290 align:middle line:84% versus going to job 2, versus the utility 00:52:33.290 --> 00:52:38.340 align:middle line:84% of MP2, which is higher than MP1 and certainly higher 00:52:38.340 --> 00:52:41.460 align:middle line:84% than MP1 minus C. So this term is clearly higher. 00:52:41.460 --> 00:52:45.380 align:middle line:84% But she doesn't get health insurance. 00:52:45.380 --> 00:52:48.500 align:middle line:84% Well, if M is someone who's really risk averse or sick, 00:52:48.500 --> 00:52:52.032 align:middle line:84% she might not make that move, even though society cares-- 00:52:52.032 --> 00:52:53.240 align:middle line:90% what does society care about? 00:52:53.240 --> 00:52:55.460 align:middle line:90% Society cares about this. 00:52:55.460 --> 00:52:58.380 align:middle line:84% Society wants people in their most productive positions. 00:52:58.380 --> 00:52:59.965 align:middle line:90% That's not happening. 00:52:59.965 --> 00:53:01.340 align:middle line:84% Emma is not moving to a job where 00:53:01.340 --> 00:53:03.007 align:middle line:84% she'd be more productive because of fear 00:53:03.007 --> 00:53:05.060 align:middle line:90% of losing health insurance. 00:53:05.060 --> 00:53:07.400 align:middle line:84% This is an economic inefficiency. 00:53:07.400 --> 00:53:08.220 align:middle line:90% And it's big. 00:53:08.220 --> 00:53:11.100 align:middle line:84% The hallmark of what makes the US so successful 00:53:11.100 --> 00:53:13.660 align:middle line:84% is the mobility of our labor market, the fact that people 00:53:13.660 --> 00:53:17.740 align:middle line:84% can seek out the most productive opportunities for themselves. 00:53:17.740 --> 00:53:20.420 align:middle line:84% When we put sand in the wheels of that system, 00:53:20.420 --> 00:53:22.080 align:middle line:90% we lower the size of the pie. 00:53:22.080 --> 00:53:22.580 align:middle line:90% Yeah? 00:53:22.580 --> 00:53:24.180 align:middle line:84% AUDIENCE: Are we to assume that she can't just 00:53:24.180 --> 00:53:25.255 align:middle line:90% purchase health care? 00:53:25.255 --> 00:53:26.380 align:middle line:90% JON GRUBER: Great question. 00:53:26.380 --> 00:53:30.300 align:middle line:84% So if she could purchase health insurance-- 00:53:30.300 --> 00:53:34.030 align:middle line:84% so you can put an upper bound on how bad job lock is. 00:53:34.030 --> 00:53:36.210 align:middle line:84% What's the upper bound on how bad job lock could be? 00:53:36.210 --> 00:53:40.150 align:middle line:90% 00:53:40.150 --> 00:53:41.330 align:middle line:90% AUDIENCE: Like in terms of-- 00:53:41.330 --> 00:53:42.038 align:middle line:90% JON GRUBER: Yeah. 00:53:42.038 --> 00:53:44.590 align:middle line:84% So you can tell me mathematically-- 00:53:44.590 --> 00:53:47.630 align:middle line:84% you can measure the upper bound on the efficiency cost of job 00:53:47.630 --> 00:53:49.310 align:middle line:90% lock. 00:53:49.310 --> 00:53:50.050 align:middle line:90% What is that? 00:53:50.050 --> 00:53:50.890 align:middle line:90% Yeah? 00:53:50.890 --> 00:53:52.230 align:middle line:84% AUDIENCE: The cost of [? private ?] insurance. 00:53:52.230 --> 00:53:52.870 align:middle line:90% JON GRUBER: What's that? 00:53:52.870 --> 00:53:53.710 align:middle line:84% AUDIENCE: The cost of [? private ?] insurance. 00:53:53.710 --> 00:53:55.830 align:middle line:84% JON GRUBER: Minus the cost of employer insurance. 00:53:55.830 --> 00:54:01.070 align:middle line:84% So basically, it's because here, she can buy insurance. 00:54:01.070 --> 00:54:06.110 align:middle line:84% So if MP1 is 20,000, M2's a million, 00:54:06.110 --> 00:54:08.550 align:middle line:84% then she just move and buy health insurance. 00:54:08.550 --> 00:54:11.750 align:middle line:84% So job lock, you might say it can't be that bad. 00:54:11.750 --> 00:54:13.750 align:middle line:84% It's just it's the gap between what 00:54:13.750 --> 00:54:15.430 align:middle line:84% your employer insurance cost, the cost of buying your own. 00:54:15.430 --> 00:54:17.388 align:middle line:84% The problem is, before the Affordable Care Act, 00:54:17.388 --> 00:54:20.057 align:middle line:84% that gap could be infinite because nongroup insurers did 00:54:20.057 --> 00:54:21.390 align:middle line:90% not have to offer you insurance. 00:54:21.390 --> 00:54:25.002 align:middle line:84% If you were sick, you were not insurable. 00:54:25.002 --> 00:54:26.210 align:middle line:90% I said I'd come back to this. 00:54:26.210 --> 00:54:27.350 align:middle line:90% But let me-- 00:54:27.350 --> 00:54:29.208 align:middle line:84% I forgot I had to cover this first. 00:54:29.208 --> 00:54:30.750 align:middle line:84% The way the nongroup insurance worked 00:54:30.750 --> 00:54:33.020 align:middle line:84% before the Affordable Care Act is, 00:54:33.020 --> 00:54:35.380 align:middle line:84% insurers were free to discriminate against you. 00:54:35.380 --> 00:54:36.980 align:middle line:84% They could not offer you insurance. 00:54:36.980 --> 00:54:39.020 align:middle line:84% They could charge you a million bucks a month. 00:54:39.020 --> 00:54:40.700 align:middle line:84% They could charge women more than men. 00:54:40.700 --> 00:54:42.400 align:middle line:84% Or they could say, for example, if you'd ever 00:54:42.400 --> 00:54:44.650 align:middle line:84% had cancer in your life and you ever get cancer again, 00:54:44.650 --> 00:54:45.840 align:middle line:90% we're not covering it. 00:54:45.840 --> 00:54:49.180 align:middle line:84% So it's possible this gap is infinite, 00:54:49.180 --> 00:54:50.840 align:middle line:84% or at least, the upper bound would be 00:54:50.840 --> 00:54:53.040 align:middle line:90% the maximum medical spending. 00:54:53.040 --> 00:54:55.660 align:middle line:84% So that's why job lock can still be bad. 00:54:55.660 --> 00:54:58.500 align:middle line:84% Now, post ACA, we can always go buy in the exchange. 00:54:58.500 --> 00:55:01.160 align:middle line:84% [INAUDIBLE] job lock should have gone down because now 00:55:01.160 --> 00:55:04.480 align:middle line:84% that upper bound is defined by what it costs you out 00:55:04.480 --> 00:55:07.560 align:middle line:84% in the world versus what your employer spends on you. 00:55:07.560 --> 00:55:12.120 align:middle line:84% Now, there was a huge literature devoted 00:55:12.120 --> 00:55:15.340 align:middle line:84% to studying this back when I was young, and I contributed to it. 00:55:15.340 --> 00:55:17.680 align:middle line:84% And the bottom line is, job lock is 00:55:17.680 --> 00:55:19.100 align:middle line:90% a real significant phenomenon. 00:55:19.100 --> 00:55:20.892 align:middle line:84% And there were studies which did two things 00:55:20.892 --> 00:55:22.880 align:middle line:84% to try to test whether job lock was real. 00:55:22.880 --> 00:55:25.400 align:middle line:84% One thing is, they looked at two employees, 00:55:25.400 --> 00:55:29.300 align:middle line:84% both at the same firm, of different health statuses, 00:55:29.300 --> 00:55:31.800 align:middle line:84% in a firm that offers health insurance and one that doesn't. 00:55:31.800 --> 00:55:33.180 align:middle line:90% So a diff in diff-- 00:55:33.180 --> 00:55:35.700 align:middle line:84% sick versus healthy in the firm that offers health insurance 00:55:35.700 --> 00:55:36.900 align:middle line:84% and sick versus healthy in the firm 00:55:36.900 --> 00:55:38.608 align:middle line:84% doesn't offer health insurance and looked 00:55:38.608 --> 00:55:40.460 align:middle line:90% at their odds of changing jobs. 00:55:40.460 --> 00:55:42.020 align:middle line:84% And what they found was, in the firm 00:55:42.020 --> 00:55:43.940 align:middle line:84% that offered health insurance, sick people 00:55:43.940 --> 00:55:45.493 align:middle line:84% were much less likely to change jobs 00:55:45.493 --> 00:55:47.660 align:middle line:84% than in the firm that didn't offer health insurance, 00:55:47.660 --> 00:55:50.500 align:middle line:84% relative to their healthy counterparts. 00:55:50.500 --> 00:55:54.020 align:middle line:84% Another is to say, look at state laws, which are called 00:55:54.020 --> 00:55:55.230 align:middle line:90% COBRA laws. 00:55:55.230 --> 00:55:56.480 align:middle line:90% You might have heard of COBRA. 00:55:56.480 --> 00:55:57.740 align:middle line:84% It's a law which allows you to continue 00:55:57.740 --> 00:55:59.365 align:middle line:84% to buy your health insurance, even when 00:55:59.365 --> 00:56:00.440 align:middle line:90% you leave your employer. 00:56:00.440 --> 00:56:02.817 align:middle line:84% What we found is, when states passed those laws, 00:56:02.817 --> 00:56:04.900 align:middle line:84% people were much more willing to leave their jobs, 00:56:04.900 --> 00:56:06.733 align:middle line:84% once again, also indicating health insurance 00:56:06.733 --> 00:56:10.340 align:middle line:90% was a barrier to job mobility. 00:56:10.340 --> 00:56:15.340 align:middle line:84% So this is suggesting this is a real potential economic 00:56:15.340 --> 00:56:18.420 align:middle line:84% inefficiency and maybe a reason why universal coverage could 00:56:18.420 --> 00:56:21.660 align:middle line:90% be economically efficient. 00:56:21.660 --> 00:56:23.580 align:middle line:90% Questions about that? 00:56:23.580 --> 00:56:24.500 align:middle line:90% Yeah? 00:56:24.500 --> 00:56:26.180 align:middle line:84% AUDIENCE: During COVID, a lot of people 00:56:26.180 --> 00:56:30.270 align:middle line:90% lost their jobs because COVID. 00:56:30.270 --> 00:56:33.150 align:middle line:84% But then if they got sick, they would not 00:56:33.150 --> 00:56:34.710 align:middle line:90% have insurance anymore? 00:56:34.710 --> 00:56:36.910 align:middle line:84% JON GRUBER: During COVID, we basically made 00:56:36.910 --> 00:56:37.993 align:middle line:90% insurance super available. 00:56:37.993 --> 00:56:39.493 align:middle line:84% So what we did is, first of all, you 00:56:39.493 --> 00:56:41.270 align:middle line:84% can always buy in the ACA exchanges. 00:56:41.270 --> 00:56:42.850 align:middle line:84% And that is dependent on your income. 00:56:42.850 --> 00:56:44.470 align:middle line:84% So if you lose your job, it's free. 00:56:44.470 --> 00:56:46.790 align:middle line:84% We also made basically anyone who wanted to eligible 00:56:46.790 --> 00:56:47.870 align:middle line:90% for Medicaid. 00:56:47.870 --> 00:56:50.010 align:middle line:84% So basically, we had a huge expansion in insurance. 00:56:50.010 --> 00:56:52.427 align:middle line:84% Insurance coverage didn't fall nearly as much during COVID 00:56:52.427 --> 00:56:54.190 align:middle line:90% as we feared. 00:56:54.190 --> 00:56:54.870 align:middle line:90% Yeah? 00:56:54.870 --> 00:56:58.250 align:middle line:84% AUDIENCE: So during COVID, we had like a single payer 00:56:58.250 --> 00:56:58.750 align:middle line:90% [INAUDIBLE]? 00:56:58.750 --> 00:56:59.610 align:middle line:90% JON GRUBER: Well, not really. 00:56:59.610 --> 00:57:00.810 align:middle line:84% And we still had the private system, 00:57:00.810 --> 00:57:02.227 align:middle line:84% but we had a lot looser conditions 00:57:02.227 --> 00:57:03.430 align:middle line:90% on getting public insurance. 00:57:03.430 --> 00:57:05.870 align:middle line:90% Yeah? 00:57:05.870 --> 00:57:08.850 align:middle line:84% AUDIENCE: The state eligibility, is it just determined by income, 00:57:08.850 --> 00:57:11.955 align:middle line:84% or is it also like, generally, the wealth that a person has? 00:57:11.955 --> 00:57:13.330 align:middle line:84% JON GRUBER: No, it's just income. 00:57:13.330 --> 00:57:16.030 align:middle line:84% Wealth comes into getting longterm care for Medicaid, 00:57:16.030 --> 00:57:17.010 align:middle line:90% like a nursing home. 00:57:17.010 --> 00:57:19.270 align:middle line:84% But for health insurance, it's just your income. 00:57:19.270 --> 00:57:22.430 align:middle line:84% AUDIENCE: So if someone is unemployed, they're [INAUDIBLE]. 00:57:22.430 --> 00:57:23.490 align:middle line:90% JON GRUBER: Yeah, yeah. 00:57:23.490 --> 00:57:25.770 align:middle line:84% So if a super rich person is getting no income, 00:57:25.770 --> 00:57:29.080 align:middle line:90% then they're eligible. 00:57:29.080 --> 00:57:29.840 align:middle line:90% OK. 00:57:29.840 --> 00:57:33.160 align:middle line:84% So last big topic I want to cover today, which is-- 00:57:33.160 --> 00:57:35.232 align:middle line:84% that's a huge amount of background. 00:57:35.232 --> 00:57:36.940 align:middle line:84% Let's now get into more of the economics. 00:57:36.940 --> 00:57:39.690 align:middle line:84% And let's start by applying the lesson of chapter 12 00:57:39.690 --> 00:57:41.542 align:middle line:90% in two ways to health care. 00:57:41.542 --> 00:57:44.000 align:middle line:84% The first way we're going to apply the lesson of chapter 12 00:57:44.000 --> 00:57:48.500 align:middle line:84% is to ask, how generous should insurance be to individuals? 00:57:48.500 --> 00:57:51.000 align:middle line:84% The second way we're going to apply the lesson of chapter 12 00:57:51.000 --> 00:57:53.800 align:middle line:84% is, how generous should insurance be to providers, 00:57:53.800 --> 00:57:55.050 align:middle line:90% in terms of reimbursement? 00:57:55.050 --> 00:57:56.300 align:middle line:90% We now have a two-sided thing. 00:57:56.300 --> 00:57:58.320 align:middle line:84% There's both, how much money should we give people? 00:57:58.320 --> 00:57:59.760 align:middle line:84% And how much money should we pay the providers 00:57:59.760 --> 00:58:00.840 align:middle line:90% who deliver the care? 00:58:00.840 --> 00:58:03.000 align:middle line:90% We'll cover the first one today. 00:58:03.000 --> 00:58:06.920 align:middle line:84% So what we're going to ask is-- let's think about a simple model 00:58:06.920 --> 00:58:10.640 align:middle line:84% where what determines the generosity of insurance-- 00:58:10.640 --> 00:58:12.960 align:middle line:84% determined by one parameter, which is how much I ask 00:58:12.960 --> 00:58:15.480 align:middle line:90% you to pay, your cost sharing. 00:58:15.480 --> 00:58:17.520 align:middle line:84% Let's take everything else out of the equation. 00:58:17.520 --> 00:58:18.780 align:middle line:90% It's comprehensive coverage. 00:58:18.780 --> 00:58:19.665 align:middle line:90% Everything's covered. 00:58:19.665 --> 00:58:21.040 align:middle line:84% The only thing that really varies 00:58:21.040 --> 00:58:24.120 align:middle line:90% is how much I make you pay. 00:58:24.120 --> 00:58:27.130 align:middle line:84% So the question is, what is the optimal amount 00:58:27.130 --> 00:58:29.450 align:middle line:90% to make someone pay? 00:58:29.450 --> 00:58:32.450 align:middle line:84% Well, here, we go back to chapter 12. 00:58:32.450 --> 00:58:34.690 align:middle line:84% And we ask, how much consumption smoothing 00:58:34.690 --> 00:58:40.205 align:middle line:84% does it do versus how much moral hazard does it cause? 00:58:40.205 --> 00:58:42.330 align:middle line:84% Well, a key insight that is developed by the famous 00:58:42.330 --> 00:58:47.650 align:middle line:84% economist Martin Feldstein in 1970s was that, in fact, 00:58:47.650 --> 00:58:49.270 align:middle line:84% if you look at insurance and ask, 00:58:49.270 --> 00:58:52.250 align:middle line:84% should insurance cover all your spending? 00:58:52.250 --> 00:58:54.730 align:middle line:84% The answer is almost certainly no 00:58:54.730 --> 00:58:56.850 align:middle line:84% because the consumption smoothing's small, 00:58:56.850 --> 00:58:58.570 align:middle line:90% and the moral hazard's big. 00:58:58.570 --> 00:59:00.370 align:middle line:84% Let's start with the consumption smoothing. 00:59:00.370 --> 00:59:03.170 align:middle line:84% Let's say you have insurance where, 00:59:03.170 --> 00:59:06.810 align:middle line:84% as much care as you use in a year, you pay $100. 00:59:06.810 --> 00:59:08.770 align:middle line:90% And I want to lower that to 0. 00:59:08.770 --> 00:59:11.930 align:middle line:90% And your job's making $60,000. 00:59:11.930 --> 00:59:14.550 align:middle line:84% That's worth to you almost exactly $100. 00:59:14.550 --> 00:59:16.453 align:middle line:84% There's no insurance value there. 00:59:16.453 --> 00:59:18.870 align:middle line:84% There's not a lot of consumption smoothing going on there. 00:59:18.870 --> 00:59:20.245 align:middle line:84% You basically pay 100 in premiums 00:59:20.245 --> 00:59:22.970 align:middle line:90% to get 100 in protection. 00:59:22.970 --> 00:59:25.560 align:middle line:84% It comes to your earlier question. 00:59:25.560 --> 00:59:28.140 align:middle line:84% If the risk is small relative to your income, 00:59:28.140 --> 00:59:31.220 align:middle line:84% there's not a lot of utility of consumption smoothing. 00:59:31.220 --> 00:59:33.157 align:middle line:84% There's not a lot of benefit of insurance. 00:59:33.157 --> 00:59:34.740 align:middle line:84% Insurance benefit comes to protect you 00:59:34.740 --> 00:59:36.500 align:middle line:84% against big risks, where you're going 00:59:36.500 --> 00:59:39.640 align:middle line:84% to really hit the curvature of that utility function. 00:59:39.640 --> 00:59:41.500 align:middle line:84% If it's a tiny risk, you're risk neutral, 00:59:41.500 --> 00:59:43.900 align:middle line:84% insurance doesn't do a lot of good for you. 00:59:43.900 --> 00:59:46.700 align:middle line:84% So that says that the consumption smoothing 00:59:46.700 --> 00:59:49.220 align:middle line:90% will be small-- 00:59:49.220 --> 00:59:53.740 align:middle line:84% remember, what determines the size of consumption smoothing? 00:59:53.740 --> 00:59:55.660 align:middle line:84% What determines how much consumption smoothing 00:59:55.660 --> 00:59:57.940 align:middle line:84% is delivered, or vice versa, how much crowd out, 00:59:57.940 --> 01:00:01.912 align:middle line:84% is if the event is unpredictable and large. 01:00:01.912 --> 01:00:04.620 align:middle line:84% Well, if I know that every year, the most I could spend in health 01:00:04.620 --> 01:00:07.180 align:middle line:84% care is $10, it's predictable I'm 01:00:07.180 --> 01:00:10.100 align:middle line:84% going to spend $10 because we all get sick sometime. 01:00:10.100 --> 01:00:11.180 align:middle line:90% And it's small. 01:00:11.180 --> 01:00:13.340 align:middle line:84% So there's no consumption smoothing. 01:00:13.340 --> 01:00:16.700 align:middle line:90% What about moral hazard? 01:00:16.700 --> 01:00:20.460 align:middle line:84% Well, Feldstein did this famous little chart we see in figure 01:00:20.460 --> 01:00:22.590 align:middle line:90% 15-7. 01:00:22.590 --> 01:00:24.910 align:middle line:84% He said, look, imagine we're going 01:00:24.910 --> 01:00:27.070 align:middle line:90% to think about doctor's visits. 01:00:27.070 --> 01:00:30.398 align:middle line:84% And imagine doctor's visits have a marginal cost of $100. 01:00:30.398 --> 01:00:32.690 align:middle line:84% That is, the social marginal cost of delivering doctors 01:00:32.690 --> 01:00:35.030 align:middle line:90% is $100. 01:00:35.030 --> 01:00:39.790 align:middle line:84% Imagine further that you're in a health insurance plan 01:00:39.790 --> 01:00:42.890 align:middle line:84% where they only make you pay $10 when you go to the doctor. 01:00:42.890 --> 01:00:45.230 align:middle line:90% That's your copayment. 01:00:45.230 --> 01:00:49.870 align:middle line:84% And then finally, let's imagine the demand for health care 01:00:49.870 --> 01:00:53.830 align:middle line:84% is downward sloping, that basically, 01:00:53.830 --> 01:00:56.390 align:middle line:84% as the price of health care goes up, you want less of it, 01:00:56.390 --> 01:00:59.750 align:middle line:90% like any other good. 01:00:59.750 --> 01:01:05.430 align:middle line:84% In that case, the optimal amount of health care you should get 01:01:05.430 --> 01:01:11.390 align:middle line:84% is at point A because the social marginal benefit of that care 01:01:11.390 --> 01:01:14.150 align:middle line:84% is dictated by your demand curve. 01:01:14.150 --> 01:01:20.228 align:middle line:84% The social marginal cost is dictated by the supply curve. 01:01:20.228 --> 01:01:22.520 align:middle line:84% And so the optimal amount of health care you should get 01:01:22.520 --> 01:01:25.040 align:middle line:90% is point A, Q1. 01:01:25.040 --> 01:01:28.480 align:middle line:90% But in fact, you choose Q2. 01:01:28.480 --> 01:01:29.580 align:middle line:90% Why do you choose Q2? 01:01:29.580 --> 01:01:32.600 align:middle line:84% You choose Q2 because that's where your private benefits 01:01:32.600 --> 01:01:35.000 align:middle line:90% equal your private costs. 01:01:35.000 --> 01:01:36.820 align:middle line:84% And that is too much health care. 01:01:36.820 --> 01:01:39.600 align:middle line:84% And indeed, it leads to a somewhat odd-looking deadweight 01:01:39.600 --> 01:01:40.840 align:middle line:90% loss triangle. 01:01:40.840 --> 01:01:44.240 align:middle line:84% But remember, deadweight loss is where costs are above benefits. 01:01:44.240 --> 01:01:47.800 align:middle line:84% And that's everything to the right of Q1 01:01:47.800 --> 01:01:49.600 align:middle line:90% and to the left of Q2. 01:01:49.600 --> 01:01:52.560 align:middle line:84% So that becomes your deadweight loss triangle. 01:01:52.560 --> 01:02:01.760 align:middle line:84% So Feldstein pointed out that by making health insurance cheap, 01:02:01.760 --> 01:02:05.240 align:middle line:90% we induce moral hazard. 01:02:05.240 --> 01:02:13.000 align:middle line:84% And what he pointed out is that basically, what that means is-- 01:02:13.000 --> 01:02:14.500 align:middle line:90% we induce moral hazard. 01:02:14.500 --> 01:02:17.720 align:middle line:84% So basically, this moral hazard problem 01:02:17.720 --> 01:02:20.433 align:middle line:84% has led people to claim that we overuse health care. 01:02:20.433 --> 01:02:23.100 align:middle line:84% And it's led people to refer to something that's called the flat 01:02:23.100 --> 01:02:25.000 align:middle line:84% of the curve, which you see in 15-8. 01:02:25.000 --> 01:02:26.400 align:middle line:90% 15-8's is a weird figure. 01:02:26.400 --> 01:02:29.840 align:middle line:90% Let me explain it. 01:02:29.840 --> 01:02:33.860 align:middle line:84% On the y-axis is the dollar of benefit 01:02:33.860 --> 01:02:38.420 align:middle line:84% you get from the next dollar of healthcare spending. 01:02:38.420 --> 01:02:42.100 align:middle line:84% On the x-axis is how much you're currently spending. 01:02:42.100 --> 01:02:47.820 align:middle line:84% So point A says, for someone currently spending $1,000, 01:02:47.820 --> 01:02:53.140 align:middle line:84% the next dollar of spending delivers $5 of benefits. 01:02:53.140 --> 01:02:57.380 align:middle line:84% For someone currently spending $2,000, 01:02:57.380 --> 01:03:01.100 align:middle line:84% the next dollar of spending delivers $1 of benefits. 01:03:01.100 --> 01:03:04.420 align:middle line:84% People understand how to interpret that graph? 01:03:04.420 --> 01:03:07.580 align:middle line:84% We think the graph for medical care production 01:03:07.580 --> 01:03:10.300 align:middle line:84% theoretically looks like this, where at first, health care's 01:03:10.300 --> 01:03:12.190 align:middle line:84% is incredibly productive, that if you 01:03:12.190 --> 01:03:13.940 align:middle line:84% go from getting no health to getting some, 01:03:13.940 --> 01:03:16.260 align:middle line:90% it's very valuable. 01:03:16.260 --> 01:03:18.030 align:middle line:84% But eventually, you reach the flat 01:03:18.030 --> 01:03:19.947 align:middle line:84% of the curve, where additional medical care is 01:03:19.947 --> 01:03:20.910 align:middle line:90% doing you no good. 01:03:20.910 --> 01:03:23.550 align:middle line:84% Indeed, some would argue it goes below zero, 01:03:23.550 --> 01:03:26.390 align:middle line:84% like the spinal fusion example, that the additional health 01:03:26.390 --> 01:03:28.470 align:middle line:84% care is actually making you worse off. 01:03:28.470 --> 01:03:29.930 align:middle line:90% But I don't need to go that far. 01:03:29.930 --> 01:03:32.790 align:middle line:90% The point is, it just gets flat. 01:03:32.790 --> 01:03:37.430 align:middle line:84% And so basically, what that says is that essentially, 01:03:37.430 --> 01:03:39.710 align:middle line:90% the situation where-- 01:03:39.710 --> 01:03:42.390 align:middle line:84% when you make health insurance free, 01:03:42.390 --> 01:03:44.510 align:middle line:84% when you make health care free, you 01:03:44.510 --> 01:03:46.610 align:middle line:84% don't do a lot of consumption smoothing, 01:03:46.610 --> 01:03:50.070 align:middle line:84% but people get a lot of health care that isn't that valuable. 01:03:50.070 --> 01:03:53.110 align:middle line:84% And that is essentially the driver 01:03:53.110 --> 01:03:56.150 align:middle line:84% of how we think about optimal health insurance, 01:03:56.150 --> 01:03:58.950 align:middle line:84% is saying that optimal health insurance would almost certainly 01:03:58.950 --> 01:04:01.310 align:middle line:90% never be free. 01:04:01.310 --> 01:04:04.910 align:middle line:84% Because on the one hand, certainly, 01:04:04.910 --> 01:04:07.870 align:middle line:84% optimal health insurance would never have you pay nothing. 01:04:07.870 --> 01:04:09.430 align:middle line:90% They'd have you pay something. 01:04:09.430 --> 01:04:10.940 align:middle line:84% But you pay nothing-- paying nothing 01:04:10.940 --> 01:04:12.690 align:middle line:84% doesn't do a lot of consumption smoothing. 01:04:12.690 --> 01:04:14.940 align:middle line:84% When I say pay, I mean what you pay at the doctor, not 01:04:14.940 --> 01:04:17.720 align:middle line:84% your premium, what you pay at the doctor, the cost sharing. 01:04:17.720 --> 01:04:20.472 align:middle line:84% Optimal cost sharing is almost certainly never zero 01:04:20.472 --> 01:04:22.680 align:middle line:84% because it doesn't do a lot of consumption smoothing, 01:04:22.680 --> 01:04:24.700 align:middle line:84% but it causes a lot of moral hazard. 01:04:24.700 --> 01:04:26.560 align:middle line:90% Yeah, question in the back? 01:04:26.560 --> 01:04:29.328 align:middle line:84% AUDIENCE: Is there a difference [? ranked ?] by the age? 01:04:29.328 --> 01:04:30.120 align:middle line:90% JON GRUBER: By age? 01:04:30.120 --> 01:04:30.745 align:middle line:90% AUDIENCE: Yeah. 01:04:30.745 --> 01:04:34.040 align:middle line:84% For example, I don't know if consumption smoothing when 01:04:34.040 --> 01:04:41.122 align:middle line:84% insurance is free, it's lower for people that are in the 60s. 01:04:41.122 --> 01:04:43.080 align:middle line:84% JON GRUBER: No, the consumption-smoothing piece 01:04:43.080 --> 01:04:45.465 align:middle line:90% will only vary by income. 01:04:45.465 --> 01:04:46.840 align:middle line:84% You might ask if the moral hazard 01:04:46.840 --> 01:04:50.600 align:middle line:84% piece varies by age because the smoothing piece is about income. 01:04:50.600 --> 01:04:53.500 align:middle line:84% If you're a millionaire, at 25 millionaire at 75, 01:04:53.500 --> 01:04:56.325 align:middle line:84% either way, $10 is irrelevant to you. 01:04:56.325 --> 01:04:58.700 align:middle line:84% So the consumption-smoothing piece doesn't depend on age. 01:04:58.700 --> 01:05:01.040 align:middle line:90% The moral hazard piece might. 01:05:01.040 --> 01:05:01.580 align:middle line:90% Yeah? 01:05:01.580 --> 01:05:04.080 align:middle line:84% AUDIENCE: Would the reason that there's an overconsumption 01:05:04.080 --> 01:05:08.360 align:middle line:84% of health care in 15-7 be different if this were 01:05:08.360 --> 01:05:11.160 align:middle line:84% an individual who was very, very close to poverty, 01:05:11.160 --> 01:05:16.168 align:middle line:84% where if they had more money, the intersection of their demand 01:05:16.168 --> 01:05:18.210 align:middle line:84% curve with the supply curve wouldn't [INAUDIBLE]? 01:05:18.210 --> 01:05:20.710 align:middle line:84% JON GRUBER: So once again, we're only doing efficiency here, 01:05:20.710 --> 01:05:21.610 align:middle line:90% not distribution. 01:05:21.610 --> 01:05:24.410 align:middle line:84% But let me ask the question differently, which is, 01:05:24.410 --> 01:05:29.090 align:middle line:84% in figure 15-7, what is the key driver of the size 01:05:29.090 --> 01:05:30.110 align:middle line:90% of the deadweight loss? 01:05:30.110 --> 01:05:32.568 align:middle line:84% There's two key drivers of the size of the deadweight loss. 01:05:32.568 --> 01:05:35.070 align:middle line:84% What drives the size of the deadweight loss in figure 15-7? 01:05:35.070 --> 01:05:35.570 align:middle line:90% Yeah? 01:05:35.570 --> 01:05:36.850 align:middle line:90% AUDIENCE: [INAUDIBLE] demand. 01:05:36.850 --> 01:05:37.558 align:middle line:90% JON GRUBER: Yeah. 01:05:37.558 --> 01:05:42.030 align:middle line:84% Well, it's the cost sharing, the gap between the two lines. 01:05:42.030 --> 01:05:43.630 align:middle line:84% But economically, more relevantly, 01:05:43.630 --> 01:05:44.505 align:middle line:90% you're exactly right. 01:05:44.505 --> 01:05:46.690 align:middle line:84% It's the elasticity of demand for health care. 01:05:46.690 --> 01:05:49.890 align:middle line:84% Another version of Steven's question is-- look, 01:05:49.890 --> 01:05:53.050 align:middle line:84% maybe healthcare demand is inelastic, especially 01:05:53.050 --> 01:05:53.910 align:middle line:90% for a per person. 01:05:53.910 --> 01:05:55.830 align:middle line:84% Maybe they only get health care if they really need it. 01:05:55.830 --> 01:05:57.663 align:middle line:84% And in some sense, it's not about the money. 01:05:57.663 --> 01:05:59.690 align:middle line:84% If you think about it, it's kind of a silly view 01:05:59.690 --> 01:06:02.790 align:middle line:84% of the world at one level, which is like, well, gee, 01:06:02.790 --> 01:06:04.610 align:middle line:84% people go to the doctor when they're sick. 01:06:04.610 --> 01:06:05.990 align:middle line:90% I don't say, that's a good deal. 01:06:05.990 --> 01:06:07.010 align:middle line:90% I'll go today. 01:06:07.010 --> 01:06:08.530 align:middle line:90% You go when you're sick. 01:06:08.530 --> 01:06:12.780 align:middle line:84% And that was the standard view of the world before the 1970s, 01:06:12.780 --> 01:06:15.900 align:middle line:84% was that basically, healthcare demand's pretty inelastic. 01:06:15.900 --> 01:06:18.320 align:middle line:84% And as healthcare demand's approximately inelastic, 01:06:18.320 --> 01:06:21.840 align:middle line:84% this deadweight loss goes to 0 because the curve gets vertical. 01:06:21.840 --> 01:06:24.260 align:middle line:84% So we don't have to worry about this. 01:06:24.260 --> 01:06:28.060 align:middle line:84% So the elasticity demand becomes the key determinant. 01:06:28.060 --> 01:06:29.780 align:middle line:84% So there became an enormous field 01:06:29.780 --> 01:06:33.420 align:middle line:84% of health economics trying to measure the elasticity of demand 01:06:33.420 --> 01:06:35.580 align:middle line:90% of health care. 01:06:35.580 --> 01:06:38.520 align:middle line:84% So initially, what studies did is, they said, 01:06:38.520 --> 01:06:42.700 align:middle line:90% well, let's compare me to Ahmet. 01:06:42.700 --> 01:06:45.920 align:middle line:84% I've got a plan with a $2,000 deductible. 01:06:45.920 --> 01:06:47.500 align:middle line:84% So I pay a lot for my health care. 01:06:47.500 --> 01:06:49.580 align:middle line:84% Ahmet's got a plan with no deductible. 01:06:49.580 --> 01:06:51.360 align:middle line:84% He doesn't pay much for his health care. 01:06:51.360 --> 01:06:53.620 align:middle line:90% Let's compare how much we spend. 01:06:53.620 --> 01:06:56.700 align:middle line:84% And what will be wrong with that? 01:06:56.700 --> 01:06:58.985 align:middle line:90% Why is that study not causal? 01:06:58.985 --> 01:07:00.860 align:middle line:84% What's the potential challenge to that study? 01:07:00.860 --> 01:07:01.800 align:middle line:90% I want to compare two people. 01:07:01.800 --> 01:07:02.020 align:middle line:90% Yeah? 01:07:02.020 --> 01:07:02.870 align:middle line:90% AUDIENCE: Is there a bias? 01:07:02.870 --> 01:07:03.620 align:middle line:90% JON GRUBER: What? 01:07:03.620 --> 01:07:05.220 align:middle line:84% AUDIENCE: The two groups of you guys will be different. 01:07:05.220 --> 01:07:07.780 align:middle line:84% JON GRUBER: In particular, how might they be different? 01:07:07.780 --> 01:07:10.002 align:middle line:84% AUDIENCE: The person who wants a lower deductible 01:07:10.002 --> 01:07:11.710 align:middle line:84% might be more health seeking [INAUDIBLE]. 01:07:11.710 --> 01:07:12.793 align:middle line:90% JON GRUBER: Yeah, exactly. 01:07:12.793 --> 01:07:15.290 align:middle line:84% So basically, you have to ask yourself-- 01:07:15.290 --> 01:07:18.110 align:middle line:84% Ahmet didn't get assigned that policy. 01:07:18.110 --> 01:07:19.007 align:middle line:90% He chose. 01:07:19.007 --> 01:07:20.090 align:middle line:90% And I didn't get assigned. 01:07:20.090 --> 01:07:20.530 align:middle line:90% I chose. 01:07:20.530 --> 01:07:22.610 align:middle line:84% Why did I choose a policy where I pay a lot out of pocket, 01:07:22.610 --> 01:07:22.970 align:middle line:90% he didn't? 01:07:22.970 --> 01:07:24.470 align:middle line:84% Presumably, because he's sicker than 01:07:24.470 --> 01:07:27.413 align:middle line:84% me and he didn't want to have to pay out of pocket. 01:07:27.413 --> 01:07:29.830 align:middle line:84% So we're going to look at-- we're going to say, wow, check 01:07:29.830 --> 01:07:30.370 align:middle line:90% it out. 01:07:30.370 --> 01:07:31.570 align:middle line:90% John has a high deductible. 01:07:31.570 --> 01:07:32.570 align:middle line:90% He spends less. 01:07:32.570 --> 01:07:34.750 align:middle line:84% Therefore, health care must be really elastic. 01:07:34.750 --> 01:07:36.230 align:middle line:90% Well, that'd be incorrect. 01:07:36.230 --> 01:07:40.470 align:middle line:90% 01:07:40.470 --> 01:07:42.450 align:middle line:84% I spend less not because of my high deductible, 01:07:42.450 --> 01:07:44.890 align:middle line:84% because people who spend less take high-deductible plans. 01:07:44.890 --> 01:07:48.070 align:middle line:84% It's classic bias, as talked about in chapter 3. 01:07:48.070 --> 01:07:49.910 align:middle line:90% So, how do you solve this? 01:07:49.910 --> 01:07:52.710 align:middle line:84% Well, as we said, the perfect solution 01:07:52.710 --> 01:07:54.470 align:middle line:84% is a randomized controlled trial. 01:07:54.470 --> 01:07:56.030 align:middle line:90% And actually, we ran one. 01:07:56.030 --> 01:07:58.470 align:middle line:84% We ran perhaps the most famous randomized controlled trial 01:07:58.470 --> 01:08:01.950 align:middle line:84% in social policy history in the US, which was the RAND Health 01:08:01.950 --> 01:08:04.830 align:middle line:90% Insurance Experiment. 01:08:04.830 --> 01:08:09.030 align:middle line:84% And what the Rand HIE did is, they literally randomized-- 01:08:09.030 --> 01:08:10.530 align:middle line:84% RAND isn't because of randomization. 01:08:10.530 --> 01:08:11.863 align:middle line:90% RAND is the name of the company. 01:08:11.863 --> 01:08:14.070 align:middle line:84% But they literally randomized health insurance. 01:08:14.070 --> 01:08:16.361 align:middle line:84% They literally took people, flipped a coin, 01:08:16.361 --> 01:08:18.569 align:middle line:84% and assigned them to three different health insurance 01:08:18.569 --> 01:08:22.210 align:middle line:84% plans, one where it was all free, one where you paid some, 01:08:22.210 --> 01:08:24.729 align:middle line:84% and one where you paid a huge amount. 01:08:24.729 --> 01:08:27.310 align:middle line:84% Essentially, think of it as comparing free insurance, 01:08:27.310 --> 01:08:30.250 align:middle line:84% where you paid nothing, to one where you had a $5,000 01:08:30.250 --> 01:08:31.010 align:middle line:90% deductible. 01:08:31.010 --> 01:08:33.210 align:middle line:84% That was sort of the comparison they did. 01:08:33.210 --> 01:08:35.490 align:middle line:90% So everyone was insured. 01:08:35.490 --> 01:08:37.865 align:middle line:84% No one had to pay more than $5,000. 01:08:37.865 --> 01:08:40.490 align:middle line:84% But the difference was, how does it matter if you pay the first 01:08:40.490 --> 01:08:44.450 align:middle line:90% $5,000 versus if you don't? 01:08:44.450 --> 01:08:47.890 align:middle line:84% Now, you might ask yourself, little side note, 01:08:47.890 --> 01:08:51.970 align:middle line:84% why would you be in this experiment if you could be 01:08:51.970 --> 01:08:53.970 align:middle line:84% assigned to a group where you have to pay $5,000 01:08:53.970 --> 01:08:56.609 align:middle line:90% for your insurance? 01:08:56.609 --> 01:08:59.111 align:middle line:90% And how is that even ethical? 01:08:59.111 --> 01:09:00.069 align:middle line:90% How do you set that up? 01:09:00.069 --> 01:09:01.736 align:middle line:84% How do you do it so it's ethical and you 01:09:01.736 --> 01:09:02.830 align:middle line:90% get people to participate? 01:09:02.830 --> 01:09:03.330 align:middle line:90% Yeah? 01:09:03.330 --> 01:09:04.770 align:middle line:84% AUDIENCE: If it's random, you also 01:09:04.770 --> 01:09:07.002 align:middle line:84% have the same chance of being [INAUDIBLE]. 01:09:07.002 --> 01:09:07.710 align:middle line:90% JON GRUBER: Yeah. 01:09:07.710 --> 01:09:10.540 align:middle line:84% But that means, on average, it's a $2,500 loss. 01:09:10.540 --> 01:09:11.399 align:middle line:90% It's random. 01:09:11.399 --> 01:09:14.140 align:middle line:84% The 0 or 5,000-- on average, you're still screwed. 01:09:14.140 --> 01:09:15.000 align:middle line:90% So what do they do? 01:09:15.000 --> 01:09:15.500 align:middle line:90% Yeah? 01:09:15.500 --> 01:09:16.720 align:middle line:90% AUDIENCE: [INAUDIBLE] 01:09:16.720 --> 01:09:17.720 align:middle line:90% JON GRUBER: What's that? 01:09:17.720 --> 01:09:21.187 align:middle line:84% AUDIENCE: I think people who [INAUDIBLE]. 01:09:21.187 --> 01:09:22.479 align:middle line:90% JON GRUBER: They could do that. 01:09:22.479 --> 01:09:25.580 align:middle line:84% But then you're getting a particularly biased answer. 01:09:25.580 --> 01:09:26.600 align:middle line:90% What else could you do? 01:09:26.600 --> 01:09:28.020 align:middle line:90% AUDIENCE: [INAUDIBLE] 01:09:28.020 --> 01:09:30.165 align:middle line:84% JON GRUBER: Give everyone $5,000. 01:09:30.165 --> 01:09:31.540 align:middle line:84% There's no bias to the experiment 01:09:31.540 --> 01:09:33.790 align:middle line:84% because the treatment and control are both getting it, 01:09:33.790 --> 01:09:35.340 align:middle line:90% and no one's worse off. 01:09:35.340 --> 01:09:37.160 align:middle line:84% And that's how you run an experiment. 01:09:37.160 --> 01:09:37.872 align:middle line:84% That's how you get people to participate 01:09:37.872 --> 01:09:39.240 align:middle line:90% in an experiment with some risk. 01:09:39.240 --> 01:09:40.880 align:middle line:84% You make sure that no one's worse off. 01:09:40.880 --> 01:09:42.520 align:middle line:84% Now, if you get the zero, you're better off. 01:09:42.520 --> 01:09:43.645 align:middle line:90% But that's just a transfer. 01:09:43.645 --> 01:09:45.899 align:middle line:84% That doesn't affect your decision to use health care. 01:09:45.899 --> 01:09:49.020 align:middle line:84% So, basically, we ran this experiment. 01:09:49.020 --> 01:09:52.260 align:middle line:84% They had several interesting findings. 01:09:52.260 --> 01:10:00.100 align:middle line:84% The first finding was that the healthcare demand 01:10:00.100 --> 01:10:02.560 align:middle line:84% curve is downward sloping, although it's not very elastic. 01:10:02.560 --> 01:10:06.670 align:middle line:84% They estimate elasticity at minus 0.2. 01:10:06.670 --> 01:10:09.050 align:middle line:84% That's for every 10% increase in healthcare prices, 01:10:09.050 --> 01:10:11.270 align:middle line:84% people consume 2% less health care. 01:10:11.270 --> 01:10:14.330 align:middle line:84% So there is deadweight loss, it is downward sloping, 01:10:14.330 --> 01:10:16.990 align:middle line:90% but it's not super steep. 01:10:16.990 --> 01:10:19.310 align:middle line:90% That was the first finding. 01:10:19.310 --> 01:10:23.670 align:middle line:84% The second finding was that basically, we 01:10:23.670 --> 01:10:27.232 align:middle line:84% had a situation where healthcare spending, 01:10:27.232 --> 01:10:29.190 align:middle line:84% the different categories of healthcare spending 01:10:29.190 --> 01:10:31.630 align:middle line:84% lined up like you think they should. 01:10:31.630 --> 01:10:34.590 align:middle line:84% What do you think should be most elastic, 01:10:34.590 --> 01:10:36.910 align:middle line:84% and what should be most inelastic? 01:10:36.910 --> 01:10:38.830 align:middle line:84% What health services should be most elastic, 01:10:38.830 --> 01:10:42.750 align:middle line:84% and what should be most inelastic? 01:10:42.750 --> 01:10:44.210 align:middle line:84% Let's get some more folks involved. 01:10:44.210 --> 01:10:44.490 align:middle line:90% Come on. 01:10:44.490 --> 01:10:45.448 align:middle line:90% This is just intuition. 01:10:45.448 --> 01:10:47.490 align:middle line:84% Let's get some of you quiet folks involved. 01:10:47.490 --> 01:10:48.010 align:middle line:90% Come on. 01:10:48.010 --> 01:10:49.370 align:middle line:84% What should be most sensitive to price? 01:10:49.370 --> 01:10:51.037 align:middle line:84% What should be least sensitive to price? 01:10:51.037 --> 01:10:54.430 align:middle line:84% You don't need to be a rocket scientist to figure this out. 01:10:54.430 --> 01:10:58.150 align:middle line:84% What services are you going to most avoid if the price goes up? 01:10:58.150 --> 01:11:00.850 align:middle line:84% And which ones are you going to not avoid if the price goes up? 01:11:00.850 --> 01:11:03.407 align:middle line:90% 01:11:03.407 --> 01:11:04.490 align:middle line:90% I don't have to cold call. 01:11:04.490 --> 01:11:05.948 align:middle line:84% Let's get some folks speaking here. 01:11:05.948 --> 01:11:07.600 align:middle line:90% Come on. 01:11:07.600 --> 01:11:08.117 align:middle line:90% Yeah? 01:11:08.117 --> 01:11:10.700 align:middle line:84% AUDIENCE: You get, like, heart surgery, [? where ?] people pay 01:11:10.700 --> 01:11:11.242 align:middle line:90% [INAUDIBLE]. 01:11:11.242 --> 01:11:11.950 align:middle line:90% JON GRUBER: Yeah. 01:11:11.950 --> 01:11:14.520 align:middle line:84% Heart surgery is pretty much inelastic. 01:11:14.520 --> 01:11:17.920 align:middle line:84% You're not going to say, yeah, $20, I'd rather die. 01:11:17.920 --> 01:11:19.960 align:middle line:90% So that's a great example. 01:11:19.960 --> 01:11:21.020 align:middle line:90% What about elastic? 01:11:21.020 --> 01:11:22.562 align:middle line:84% What's something where people-- yeah? 01:11:22.562 --> 01:11:23.645 align:middle line:90% AUDIENCE: Plastic surgery. 01:11:23.645 --> 01:11:25.040 align:middle line:90% JON GRUBER: Plastic surgery. 01:11:25.040 --> 01:11:29.885 align:middle line:84% The most elastic thing-- dental care, mental health care, 01:11:29.885 --> 01:11:31.760 align:middle line:84% things which are necessary medical treatments 01:11:31.760 --> 01:11:35.400 align:middle line:84% for many people, but which people pay attention to price 01:11:35.400 --> 01:11:36.700 align:middle line:90% and are more price sensitive. 01:11:36.700 --> 01:11:38.440 align:middle line:84% And RAND insurance found exactly, 01:11:38.440 --> 01:11:43.120 align:middle line:84% that the elasticities line up exactly as you'd think. 01:11:43.120 --> 01:11:48.720 align:middle line:84% Then the most striking finding was, OK, 01:11:48.720 --> 01:11:51.160 align:middle line:90% people use less health care. 01:11:51.160 --> 01:11:53.580 align:middle line:84% They actually use less of all health care, 01:11:53.580 --> 01:11:55.520 align:middle line:90% including preventive care. 01:11:55.520 --> 01:11:59.400 align:middle line:84% So one criticism you hear of making people pay for health 01:11:59.400 --> 01:12:00.680 align:middle line:90% care is, wait a second. 01:12:00.680 --> 01:12:03.260 align:middle line:84% What if they stop using not just the stuff they don't need 01:12:03.260 --> 01:12:05.770 align:middle line:90% but the stuff they do need? 01:12:05.770 --> 01:12:08.010 align:middle line:90% And maybe their health is worse. 01:12:08.010 --> 01:12:13.650 align:middle line:84% So RAND also looked to people's health and found no effect. 01:12:13.650 --> 01:12:15.970 align:middle line:84% Then we're on the flat of the curve. 01:12:15.970 --> 01:12:17.950 align:middle line:84% People who, with $5,000 deductibles, 01:12:17.950 --> 01:12:21.170 align:middle line:84% who used a lot less health care, end up just as healthy as people 01:12:21.170 --> 01:12:22.890 align:middle line:90% who didn't. 01:12:22.890 --> 01:12:25.010 align:middle line:90% So what's going on with that? 01:12:25.010 --> 01:12:29.210 align:middle line:90% Well, there's two possibilities. 01:12:29.210 --> 01:12:31.610 align:middle line:84% One possibility is that we don't really 01:12:31.610 --> 01:12:34.130 align:middle line:84% know what effective and ineffective health is, 01:12:34.130 --> 01:12:34.950 align:middle line:90% and people do. 01:12:34.950 --> 01:12:35.830 align:middle line:90% And they just use less of it. 01:12:35.830 --> 01:12:37.330 align:middle line:84% And it turned out, it didn't matter. 01:12:37.330 --> 01:12:38.420 align:middle line:90% They just use less of it. 01:12:38.420 --> 01:12:40.170 align:middle line:84% The other possibility is, maybe the people 01:12:40.170 --> 01:12:42.010 align:middle line:84% who used less understood when they could afford 01:12:42.010 --> 01:12:44.130 align:middle line:84% to use-- the related is, people sort of understood 01:12:44.130 --> 01:12:46.690 align:middle line:84% when it was OK to use less and not use less. 01:12:46.690 --> 01:12:50.425 align:middle line:84% But that was a striking finding, that we really look like 01:12:50.425 --> 01:12:53.050 align:middle line:84% we're on the flat of the curve, that going from the 0 to $5,000 01:12:53.050 --> 01:12:55.008 align:middle line:84% deductible-- now, remember, no one's uninsured. 01:12:55.008 --> 01:12:56.370 align:middle line:90% We're not saying you pay it all. 01:12:56.370 --> 01:12:57.950 align:middle line:84% There was a cap on what you'd pay. 01:12:57.950 --> 01:12:59.855 align:middle line:84% We'll come back to the uninsured next time. 01:12:59.855 --> 01:13:01.230 align:middle line:84% There was a cap on what you paid. 01:13:01.230 --> 01:13:03.180 align:middle line:84% But at least within that range, where 01:13:03.180 --> 01:13:05.580 align:middle line:84% you are spending a little versus a lot, 01:13:05.580 --> 01:13:07.860 align:middle line:84% it's not mattering for your health, 01:13:07.860 --> 01:13:12.260 align:middle line:84% except they did find suggestive evidence of effects 01:13:12.260 --> 01:13:16.420 align:middle line:84% for low-income, very sick people, what 01:13:16.420 --> 01:13:18.980 align:middle line:84% we call chronically ill people, people with diabetes 01:13:18.980 --> 01:13:20.580 align:middle line:84% or high blood pressure or depression 01:13:20.580 --> 01:13:22.860 align:middle line:84% and other things, chronic conditions. 01:13:22.860 --> 01:13:27.620 align:middle line:84% For that group, they did find suggestion 01:13:27.620 --> 01:13:31.420 align:middle line:84% that these higher prices were bad for them. 01:13:31.420 --> 01:13:35.220 align:middle line:84% And that comes-- someone mentioned different incomes. 01:13:35.220 --> 01:13:38.300 align:middle line:84% That is a very interesting lesson 01:13:38.300 --> 01:13:42.780 align:middle line:84% for how we think about optimal health insurance, which 01:13:42.780 --> 01:13:45.780 align:middle line:84% says that basically, what RAND tells 01:13:45.780 --> 01:13:51.940 align:middle line:84% us is that if you are in a range where the money is not that 01:13:51.940 --> 01:13:54.660 align:middle line:84% meaningful to you, then probably, 01:13:54.660 --> 01:13:56.980 align:middle line:84% it doesn't really affect your health about whether you 01:13:56.980 --> 01:13:58.000 align:middle line:90% use the health care or not. 01:13:58.000 --> 01:13:59.120 align:middle line:84% If it's really going to affect your health, 01:13:59.120 --> 01:14:00.550 align:middle line:90% you'll get it anyway. 01:14:00.550 --> 01:14:04.070 align:middle line:84% But if you're poor and sick where you really just 01:14:04.070 --> 01:14:06.150 align:middle line:84% can't afford it, you might not have it, 01:14:06.150 --> 01:14:08.270 align:middle line:84% even if it could improve your health. 01:14:08.270 --> 01:14:12.590 align:middle line:84% And that leads us to a subtle explanation of optimal health 01:14:12.590 --> 01:14:16.390 align:middle line:84% insurance, which is, optimal health insurance is not just 01:14:16.390 --> 01:14:18.750 align:middle line:84% saying, well, we should have a deductible. 01:14:18.750 --> 01:14:21.050 align:middle line:84% So before I made that point-- in fact, let me go back. 01:14:21.050 --> 01:14:21.810 align:middle line:90% I did this in the wrong order. 01:14:21.810 --> 01:14:24.185 align:middle line:84% If I'd stop before that last point, if I just said, look, 01:14:24.185 --> 01:14:28.370 align:middle line:84% Rand found, $5,000 deductible, no worse health, you'd say, 01:14:28.370 --> 01:14:29.005 align:middle line:90% well, great. 01:14:29.005 --> 01:14:30.630 align:middle line:84% We should at least have a decent amount 01:14:30.630 --> 01:14:33.670 align:middle line:84% of cost sharing, not a lot of consumption smoothing, 01:14:33.670 --> 01:14:36.772 align:middle line:84% and reduce moral hazard without impacting health. 01:14:36.772 --> 01:14:38.230 align:middle line:84% But then there's this other finding 01:14:38.230 --> 01:14:41.310 align:middle line:84% that for low-income people, that may impact their health. 01:14:41.310 --> 01:14:43.470 align:middle line:84% What does that suggest we should do? 01:14:43.470 --> 01:14:47.350 align:middle line:84% That suggests that we should have income-related cost 01:14:47.350 --> 01:14:49.350 align:middle line:84% sharing, where the amount you have 01:14:49.350 --> 01:14:55.070 align:middle line:84% to pay when you go to the doctor should 01:14:55.070 --> 01:14:58.230 align:middle line:84% depend on the resource you have to pay it. 01:14:58.230 --> 01:15:02.690 align:middle line:84% In that way, you reduce moral hazard to people 01:15:02.690 --> 01:15:05.610 align:middle line:84% where you can really reduce it without sacrificing health. 01:15:05.610 --> 01:15:09.030 align:middle line:84% But the ones that will sacrifice health, 01:15:09.030 --> 01:15:11.690 align:middle line:84% you avoid them reducing their care. 01:15:11.690 --> 01:15:14.042 align:middle line:84% That would suggest maybe an income-based cost sharing. 01:15:14.042 --> 01:15:16.250 align:middle line:84% And we actually have that in the Affordable Care Act. 01:15:16.250 --> 01:15:18.083 align:middle line:84% We actually have, the higher your income is, 01:15:18.083 --> 01:15:19.690 align:middle line:84% the more you pay out of pocket, to try 01:15:19.690 --> 01:15:22.370 align:middle line:90% to address that exact concern. 01:15:22.370 --> 01:15:25.730 align:middle line:84% An income-based out-of-pocket system 01:15:25.730 --> 01:15:28.970 align:middle line:84% is one form of optimal insurance. 01:15:28.970 --> 01:15:30.930 align:middle line:84% The other piece of optimal insurance 01:15:30.930 --> 01:15:33.890 align:middle line:84% comes to something we call value-based insurance 01:15:33.890 --> 01:15:37.210 align:middle line:84% design, which is saying, look, we 01:15:37.210 --> 01:15:41.290 align:middle line:84% do know with certainty some services are high value, 01:15:41.290 --> 01:15:42.350 align:middle line:90% and some are low value. 01:15:42.350 --> 01:15:44.690 align:middle line:84% For example, if you're diabetic, there's 01:15:44.690 --> 01:15:49.410 align:middle line:84% a drug you can take which is $1 a day, which basically keeps you 01:15:49.410 --> 01:15:52.850 align:middle line:84% from getting your foot amputated, keeps you alive. 01:15:52.850 --> 01:15:53.710 align:middle line:90% That's effective. 01:15:53.710 --> 01:15:55.490 align:middle line:90% You should be doing that. 01:15:55.490 --> 01:15:58.860 align:middle line:84% On the other hand, we know this back surgery is ineffective, 01:15:58.860 --> 01:16:00.620 align:middle line:90% maybe even bad for you. 01:16:00.620 --> 01:16:03.300 align:middle line:84% So what a value-based insurance design would say is, look, 01:16:03.300 --> 01:16:11.340 align:middle line:84% we ought to vary what people pay by how effective the service is. 01:16:11.340 --> 01:16:13.100 align:middle line:84% So insurance, it's not just one cost 01:16:13.100 --> 01:16:15.420 align:middle line:84% sharing based on your income but actually varies 01:16:15.420 --> 01:16:18.120 align:middle line:84% based on how appropriate the service is for you. 01:16:18.120 --> 01:16:21.240 align:middle line:84% You could even have personalized cost sharing, where you say, 01:16:21.240 --> 01:16:22.400 align:middle line:90% well, John's a diabetic. 01:16:22.400 --> 01:16:24.860 align:middle line:84% Therefore, we're going to make this pill free for him. 01:16:24.860 --> 01:16:28.540 align:middle line:84% But his back surgery is going to be really expensive, et cetera. 01:16:28.540 --> 01:16:31.820 align:middle line:84% And that would be the ultimate way to design insurance. 01:16:31.820 --> 01:16:33.780 align:middle line:84% The problem is, it's really hard. 01:16:33.780 --> 01:16:36.420 align:middle line:84% So I was appointed-- when we passed healthcare reform 01:16:36.420 --> 01:16:38.963 align:middle line:84% in Massachusetts, the law was incredibly vague. 01:16:38.963 --> 01:16:40.880 align:middle line:84% The law said, we should have health insurance. 01:16:40.880 --> 01:16:42.380 align:middle line:90% There should be good benefits. 01:16:42.380 --> 01:16:43.240 align:middle line:90% Go figure that out. 01:16:43.240 --> 01:16:45.020 align:middle line:84% And they passed that on to a board of 10 people, of which I 01:16:45.020 --> 01:16:45.860 align:middle line:90% was one. 01:16:45.860 --> 01:16:47.940 align:middle line:84% We had to figure out what constitutes good health 01:16:47.940 --> 01:16:51.240 align:middle line:84% insurance and what people should pay for it. 01:16:51.240 --> 01:16:52.240 align:middle line:90% So I was pretty excited. 01:16:52.240 --> 01:16:54.140 align:middle line:84% I called my buddy, who came up with the term 01:16:54.140 --> 01:16:56.432 align:middle line:84% value-based insurance design, sort of the king of this. 01:16:56.432 --> 01:16:58.270 align:middle line:90% I said, this is so exciting. 01:16:58.270 --> 01:16:59.378 align:middle line:90% I'm in charge, man. 01:16:59.378 --> 01:17:01.170 align:middle line:84% We can set up value-based insurance design. 01:17:01.170 --> 01:17:02.690 align:middle line:84% What should we-- he goes, this is so exciting. 01:17:02.690 --> 01:17:04.550 align:middle line:84% He said, let's charge higher copayments 01:17:04.550 --> 01:17:06.530 align:middle line:84% for low-value services and lower copayments 01:17:06.530 --> 01:17:07.530 align:middle line:90% for high-value services. 01:17:07.530 --> 01:17:09.010 align:middle line:84% I'm like, that's awesome, I want to do that. 01:17:09.010 --> 01:17:10.250 align:middle line:90% What should we actually do? 01:17:10.250 --> 01:17:12.390 align:middle line:84% And he said, let's charge higher copayments 01:17:12.390 --> 01:17:14.310 align:middle line:84% for low-value services and lower copayments 01:17:14.310 --> 01:17:15.310 align:middle line:90% for high-value services. 01:17:15.310 --> 01:17:17.470 align:middle line:84% Because it's actually hard to figure out how 01:17:17.470 --> 01:17:19.070 align:middle line:90% you should actually do it. 01:17:19.070 --> 01:17:22.230 align:middle line:84% So the problem is-- this is a standard thing economists run 01:17:22.230 --> 01:17:25.670 align:middle line:84% into in policy design, which is, the optimal policy in theory 01:17:25.670 --> 01:17:28.390 align:middle line:84% may be actually hard to implement in practice. 01:17:28.390 --> 01:17:31.710 align:middle line:84% So where do we come, in terms of optimal health insurance? 01:17:31.710 --> 01:17:34.870 align:middle line:84% Where we come down, in terms of optimal health insurance, 01:17:34.870 --> 01:17:38.950 align:middle line:84% is that basically, zero cost sharing almost certainly 01:17:38.950 --> 01:17:41.950 align:middle line:84% doesn't make sense unless you're very poor, 01:17:41.950 --> 01:17:45.110 align:middle line:84% but that the cost sharing you pay should depend on your income 01:17:45.110 --> 01:17:48.350 align:middle line:84% and probably should depend on what the service is as well. 01:17:48.350 --> 01:17:48.850 align:middle line:90% OK. 01:17:48.850 --> 01:17:51.220 align:middle line:90% I'm going to stop there. 01:17:51.220 --> 01:17:55.000 align:middle line:90%