1 00:00:00,000 --> 00:00:02,520 The following content is provided under a Creative 2 00:00:02,520 --> 00:00:03,970 Commons license. 3 00:00:03,970 --> 00:00:06,360 Your support will help MIT OpenCourseWare 4 00:00:06,360 --> 00:00:10,660 continue to offer high quality educational resources for free. 5 00:00:10,660 --> 00:00:13,320 To make a donation or view additional materials 6 00:00:13,320 --> 00:00:17,170 from hundreds of MIT courses, visit MIT OpenCourseWare 7 00:00:17,170 --> 00:00:18,450 at ocw.mit.edu. 8 00:00:25,285 --> 00:00:26,660 RICHARD SCHMALENSEE: Today, we're 9 00:00:26,660 --> 00:00:28,150 talking about energy demand. 10 00:00:28,150 --> 00:00:31,880 I do owe you 2 pictures however, which I will now supply. 11 00:00:31,880 --> 00:00:37,400 The first is-- that's Bangladesh as it 12 00:00:37,400 --> 00:00:40,430 is today with 112 million people, 13 00:00:40,430 --> 00:00:45,410 and as it would be with a meter and a half sea level rise, 14 00:00:45,410 --> 00:00:49,910 about 17 million people or 15% of the population 15 00:00:49,910 --> 00:00:52,160 live in the areas that would be flooded. 16 00:00:52,160 --> 00:00:54,350 A meter and a half may be a little 17 00:00:54,350 --> 00:00:57,440 aggressive over the century, but it's within the bounds 18 00:00:57,440 --> 00:00:58,970 that people talk about. 19 00:00:58,970 --> 00:01:01,760 They talk about a meter, meter and a 1/2, sometimes 2. 20 00:01:01,760 --> 00:01:04,580 So that's Bangladesh. 21 00:01:04,580 --> 00:01:07,430 This if it comes out, this is a picture 22 00:01:07,430 --> 00:01:09,890 that I promised you to talk about-- 23 00:01:09,890 --> 00:01:14,590 radiative forcing and methane here versus carbon dioxide. 24 00:01:14,590 --> 00:01:21,770 So this is a 2007 study that looks at total contribution 25 00:01:21,770 --> 00:01:29,150 to increased radiative impact between 1750 and 2005. 26 00:01:29,150 --> 00:01:33,200 You will see up here, CO2 has a much bigger estimated effect 27 00:01:33,200 --> 00:01:34,700 than methane. 28 00:01:34,700 --> 00:01:40,070 And the molecules last a lot longer in the atmosphere. 29 00:01:40,070 --> 00:01:42,950 So each methane molecule, as Susan Solomon said, 30 00:01:42,950 --> 00:01:44,930 is more powerful. 31 00:01:44,930 --> 00:01:47,120 But half of these will be gone in 10 years. 32 00:01:47,120 --> 00:01:49,070 Whereas ballpark, half of these will still 33 00:01:49,070 --> 00:01:50,840 be around in a century. 34 00:01:50,840 --> 00:01:55,940 So there are clear reasons for focusing on carbon dioxide. 35 00:01:55,940 --> 00:01:59,600 The halo carbons are fluorocarbons, 36 00:01:59,600 --> 00:02:03,800 chlorofluorocarbons and the like. 37 00:02:03,800 --> 00:02:06,920 They are very powerful, and they last a long time. 38 00:02:06,920 --> 00:02:10,130 But they're not greatly abundant. 39 00:02:10,130 --> 00:02:13,310 And people are working to phase them out. 40 00:02:13,310 --> 00:02:14,750 Somebody mentioned black carbon. 41 00:02:14,750 --> 00:02:19,040 There's a black carbon impact via the reflectivity, 42 00:02:19,040 --> 00:02:20,770 the albedo. 43 00:02:20,770 --> 00:02:25,410 It does have a positive impact, land use, negative impact 44 00:02:25,410 --> 00:02:27,620 and so forth. 45 00:02:27,620 --> 00:02:30,110 Aerosols are interesting. 46 00:02:30,110 --> 00:02:33,290 This is like sulfur dioxide put in the air 47 00:02:33,290 --> 00:02:35,180 by coal-fired power plants. 48 00:02:35,180 --> 00:02:38,390 That reflects, so that tends to cool. 49 00:02:38,390 --> 00:02:41,390 When Mount Pinatubo blew in the Philippines, 50 00:02:41,390 --> 00:02:44,750 there was I think about a two degree impact on climate 51 00:02:44,750 --> 00:02:47,660 from basically all that dust reflecting sunlight. 52 00:02:47,660 --> 00:02:50,940 So that's the best estimate. 53 00:02:50,940 --> 00:02:54,260 It appears lots of places of the cumulative impact 54 00:02:54,260 --> 00:02:57,830 of human activity. 55 00:02:57,830 --> 00:03:00,860 1750 is usually taken as pre-industrial times, 56 00:03:00,860 --> 00:03:02,780 kind of the benchmark. 57 00:03:02,780 --> 00:03:04,290 Does this make some sense? 58 00:03:04,290 --> 00:03:07,820 Are there questions about it? 59 00:03:07,820 --> 00:03:09,770 OK, hearing none. 60 00:03:12,770 --> 00:03:15,710 Today, I want to talk about energy demand. 61 00:03:15,710 --> 00:03:17,570 We'll also talk about it Wednesday 62 00:03:17,570 --> 00:03:21,720 but from a rather different point of view. 63 00:03:21,720 --> 00:03:24,780 I want to talk about where it comes from, 64 00:03:24,780 --> 00:03:27,720 the sort of short run, long run difference. 65 00:03:27,720 --> 00:03:30,840 How you might estimate characteristics 66 00:03:30,840 --> 00:03:35,820 of energy demand and then stability and some puzzles. 67 00:03:35,820 --> 00:03:39,180 So this is basically going to be a walk through that teaching 68 00:03:39,180 --> 00:03:40,410 note. 69 00:03:40,410 --> 00:03:45,700 So it's fairly basic, fairly basic stuff. 70 00:03:45,700 --> 00:03:51,050 With very few exceptions, there is no demand for energy. 71 00:03:51,050 --> 00:03:53,870 There's a demand for what energy does. 72 00:03:53,870 --> 00:03:56,500 I mean you don't buy gasoline to have it around the house. 73 00:03:56,500 --> 00:03:58,060 Gasoline is nasty. 74 00:03:58,060 --> 00:03:59,890 Vapors are explosive. 75 00:03:59,890 --> 00:04:01,820 You buy it for what you can do with it. 76 00:04:04,990 --> 00:04:09,280 The demand for gasoline is derived from what 77 00:04:09,280 --> 00:04:13,220 it does, what it does for you. 78 00:04:13,220 --> 00:04:17,740 And if you think about all the things you use energy for, I 79 00:04:17,740 --> 00:04:20,110 mean, you use it to heat food. 80 00:04:20,110 --> 00:04:21,355 You use it to heat the house. 81 00:04:21,355 --> 00:04:22,690 You use it to cool. 82 00:04:22,690 --> 00:04:25,300 You use it to make loud music. 83 00:04:25,300 --> 00:04:27,738 Energy does a lot of things. 84 00:04:27,738 --> 00:04:30,280 You don't actually want to run electricity through your body, 85 00:04:30,280 --> 00:04:33,170 you want to use it to drive speakers and so forth. 86 00:04:33,170 --> 00:04:38,210 So the question is, if it's that kind of a commodity, 87 00:04:38,210 --> 00:04:40,060 how do we think about it's demand? 88 00:04:40,060 --> 00:04:42,380 Where does it come from? 89 00:04:42,380 --> 00:04:48,400 It's clear that to produce any energy service, heat, cooling, 90 00:04:48,400 --> 00:04:52,150 transportation, it comes with a bunch of inputs, 91 00:04:52,150 --> 00:04:56,320 like any other product. 92 00:04:56,320 --> 00:04:58,210 Transportation will involve capital 93 00:04:58,210 --> 00:05:00,370 in the form of some kind of vehicle, 94 00:05:00,370 --> 00:05:03,280 driving energy, in this case, not 95 00:05:03,280 --> 00:05:05,170 materials other than gasoline. 96 00:05:05,170 --> 00:05:11,760 But well, maybe spare parts, oil, and so forth. 97 00:05:11,760 --> 00:05:12,980 So you start with that. 98 00:05:16,770 --> 00:05:21,440 And I just make this interesting point 99 00:05:21,440 --> 00:05:26,060 that you normally think of inputs 100 00:05:26,060 --> 00:05:27,590 into production as substitutes. 101 00:05:27,590 --> 00:05:31,610 So if capital gets more expensive, you use more labor. 102 00:05:31,610 --> 00:05:33,150 You switch from capital to labor. 103 00:05:33,150 --> 00:05:34,820 Well, capital and energy has turned out 104 00:05:34,820 --> 00:05:37,430 in a number of studies to be complements, 105 00:05:37,430 --> 00:05:38,780 which is pretty rare. 106 00:05:38,780 --> 00:05:40,700 If you raise the price of energy, 107 00:05:40,700 --> 00:05:42,230 you reduce the demand for energy. 108 00:05:42,230 --> 00:05:44,360 You reduce the demand for capital 109 00:05:44,360 --> 00:05:47,780 used in production as well, OK. 110 00:05:47,780 --> 00:05:53,240 I will talk about this last point in a minute, the progress 111 00:05:53,240 --> 00:05:56,000 in these production functions. 112 00:05:56,000 --> 00:05:58,270 Clear so far? 113 00:05:58,270 --> 00:05:58,770 Yeah. 114 00:05:58,770 --> 00:06:00,150 AUDIENCE: Sorry, what example? 115 00:06:00,150 --> 00:06:02,120 RICHARD SCHMALENSEE: Materials. 116 00:06:02,120 --> 00:06:05,180 That's just a general form of a production function. 117 00:06:08,100 --> 00:06:10,770 If you want to produce cooked food, 118 00:06:10,770 --> 00:06:14,610 you need some piece of capital to contain 119 00:06:14,610 --> 00:06:18,280 the heat, some source of heat, your cooking time, 120 00:06:18,280 --> 00:06:19,990 and of course, the ingredients. 121 00:06:19,990 --> 00:06:21,390 So that would be an example where 122 00:06:21,390 --> 00:06:26,610 it would be important, OK? 123 00:06:26,610 --> 00:06:28,680 This is just some indication. 124 00:06:28,680 --> 00:06:32,760 There's a Nordhaus paper I think I cited in the note that 125 00:06:32,760 --> 00:06:37,830 looks at the cost of lighting. 126 00:06:37,830 --> 00:06:42,270 Cents per 1,000 lumens from 1800 to today. 127 00:06:42,270 --> 00:06:43,878 And it's dramatic. 128 00:06:43,878 --> 00:06:45,420 You can actually do light because you 129 00:06:45,420 --> 00:06:47,640 can look at the output of various lighting devices. 130 00:06:47,640 --> 00:06:50,410 You're doing whale oil up here and so forth. 131 00:06:50,410 --> 00:06:53,070 And it does get a little flat. 132 00:06:53,070 --> 00:06:55,350 Once you get the fluorescent bulbs, it gets flat. 133 00:06:55,350 --> 00:06:58,020 And we're down here beginning to think about LEDs. 134 00:06:58,020 --> 00:07:02,370 But it's also true in automobiles. 135 00:07:02,370 --> 00:07:08,565 I don't have the slide, but 2.6% of the year advance 136 00:07:08,565 --> 00:07:12,840 in automotive technology estimated by my colleague, 137 00:07:12,840 --> 00:07:16,590 Chris Knittel, means that-- 138 00:07:16,590 --> 00:07:20,530 well, here's one example of what it means. 139 00:07:20,530 --> 00:07:23,530 If you kept all the other characteristics 140 00:07:23,530 --> 00:07:26,350 of an automobile at 1980 levels, we'd 141 00:07:26,350 --> 00:07:29,380 have a 50% increase in miles per gallon, 142 00:07:29,380 --> 00:07:34,300 or decrease in gallons per mile between 1980 and 2006. 143 00:07:34,300 --> 00:07:35,950 In fact, we didn't. 144 00:07:35,950 --> 00:07:40,430 Mileage was pretty stable and all these things increased. 145 00:07:40,430 --> 00:07:44,180 But 2.6% a year is a pretty dramatic increase 146 00:07:44,180 --> 00:07:49,980 in the advance of a very mature technology. 147 00:07:49,980 --> 00:07:52,410 And you see this in a lot of energy using technology. 148 00:07:52,410 --> 00:07:57,000 Power plants have this lighting, motors, 149 00:07:57,000 --> 00:07:58,920 lots of energy using technologies, 150 00:07:58,920 --> 00:08:02,640 air conditioners have this kind of progress 151 00:08:02,640 --> 00:08:08,850 that energy per unit of output has fallen dramatically 152 00:08:08,850 --> 00:08:10,650 historically. 153 00:08:10,650 --> 00:08:15,990 All right, so let me do a little bit of characteristics 154 00:08:15,990 --> 00:08:18,000 of derived demand. 155 00:08:18,000 --> 00:08:21,510 Because when you think about energy, energy demand, you do 156 00:08:21,510 --> 00:08:24,290 need to kind think about it as derived demand. 157 00:08:24,290 --> 00:08:27,000 So let's suppose capital is fixed. 158 00:08:27,000 --> 00:08:29,400 It's the short run. 159 00:08:29,400 --> 00:08:33,230 I'm going to just lay this out for the moment. 160 00:08:33,230 --> 00:08:38,908 Yep, so we want to drive the demand for gasoline. 161 00:08:38,908 --> 00:08:40,450 We want to think about where it comes 162 00:08:40,450 --> 00:08:48,110 from, or the demand for gas to dry clothes, or to cook food. 163 00:08:48,110 --> 00:08:51,920 There is sort of two functions involved. 164 00:08:51,920 --> 00:08:55,000 The first is some demand function for the service. 165 00:08:55,000 --> 00:08:58,240 What am I willing to pay to have my condo cool? 166 00:08:58,240 --> 00:09:03,430 That's going to depend on the cost of cooling it, 167 00:09:03,430 --> 00:09:06,460 other prices, how wealthy I am, capital stock 168 00:09:06,460 --> 00:09:08,170 tastes, all kinds of things. 169 00:09:08,170 --> 00:09:14,320 But I'm going to focus on the relation between how 170 00:09:14,320 --> 00:09:17,350 cool I keep it, my demand for cooling, 171 00:09:17,350 --> 00:09:20,440 and what does it cost me to cool? 172 00:09:20,440 --> 00:09:25,410 What it costs me to cool comes out of how I produce cooling. 173 00:09:25,410 --> 00:09:29,270 There's some production function determined by my very old air 174 00:09:29,270 --> 00:09:32,900 conditioner that I'll get cooling services. 175 00:09:32,900 --> 00:09:35,330 It will depend on, well, in this case not particularly 176 00:09:35,330 --> 00:09:36,380 materials or labor. 177 00:09:36,380 --> 00:09:39,890 It's just going to depend on how much electricity I use. 178 00:09:39,890 --> 00:09:44,630 So in the case of cooling, this function gets simple. 179 00:09:44,630 --> 00:09:46,910 The cooling, however we measure it 180 00:09:46,910 --> 00:09:49,070 is going to be a function of how much energy, 181 00:09:49,070 --> 00:09:51,020 how much electricity I use. 182 00:09:51,020 --> 00:09:51,980 We could get fancy. 183 00:09:51,980 --> 00:09:54,030 It will depend on when I use it and so forth. 184 00:09:54,030 --> 00:09:57,650 But as a simple story, that's a pretty good story. 185 00:09:57,650 --> 00:10:03,510 So if what I'm interested in is the demand for electricity 186 00:10:03,510 --> 00:10:10,770 as a function of its price, then somehow I 187 00:10:10,770 --> 00:10:13,980 don't have a price for cooling, right? 188 00:10:13,980 --> 00:10:14,550 I don't pay. 189 00:10:14,550 --> 00:10:16,380 I don't get a bill to cool my apartment. 190 00:10:16,380 --> 00:10:19,060 I get a bill for electricity. 191 00:10:19,060 --> 00:10:22,110 So if I want to get a demand for electricity 192 00:10:22,110 --> 00:10:25,980 as a function of its price, I need these. 193 00:10:25,980 --> 00:10:29,170 I need this function. 194 00:10:29,170 --> 00:10:30,850 How much do I care about cooling? 195 00:10:30,850 --> 00:10:33,940 And this function, how do I get cooling? 196 00:10:33,940 --> 00:10:37,580 And then obviously, something about the price of electricity. 197 00:10:37,580 --> 00:10:41,530 So I'll tell you how you do it in general, 198 00:10:41,530 --> 00:10:43,210 and then we'll walk through an example. 199 00:10:43,210 --> 00:10:47,020 And there's some stuff in the notes. 200 00:10:47,020 --> 00:10:50,830 So if I've got all the prices, conceptually I 201 00:10:50,830 --> 00:10:55,110 say, well, what's the best way to produce cooling? 202 00:10:55,110 --> 00:11:03,450 That is a function of prices, prices of the way 203 00:11:03,450 --> 00:11:04,740 down here rather. 204 00:11:04,740 --> 00:11:08,250 If I know the prices of energy, the price of anything else I 205 00:11:08,250 --> 00:11:12,600 use, I can say, what's the cost of cooling 206 00:11:12,600 --> 00:11:15,450 as a function of various prices? 207 00:11:15,450 --> 00:11:17,730 Then I can substitute the cost of cooling, 208 00:11:17,730 --> 00:11:22,770 the price of the service up here and get the demand. 209 00:11:22,770 --> 00:11:24,970 That's impossible to follow. 210 00:11:24,970 --> 00:11:25,470 Am I right? 211 00:11:25,470 --> 00:11:27,810 OK. 212 00:11:27,810 --> 00:11:31,530 You'll come back to it after the example. 213 00:11:31,530 --> 00:11:36,460 So let's do lighting with two very simple functions. 214 00:11:36,460 --> 00:11:41,200 Suppose up here is my demand for lighting. 215 00:11:41,200 --> 00:11:45,240 So lumen hours, which I'm just going 216 00:11:45,240 --> 00:11:47,640 to call lumens to keep my slides a little shorter, 217 00:11:47,640 --> 00:11:49,770 my bullet's a little shorter. 218 00:11:49,770 --> 00:11:52,350 Lumen hours, this is my demand. 219 00:11:52,350 --> 00:11:55,020 If this is the cost per lumen hour, 220 00:11:55,020 --> 00:11:58,230 then with A and B as constants, then this 221 00:11:58,230 --> 00:11:59,670 is my demand per lumen hour. 222 00:11:59,670 --> 00:12:03,360 It's a straightforward demand curve. 223 00:12:03,360 --> 00:12:06,030 B is the price elasticity. 224 00:12:06,030 --> 00:12:10,020 You may remember that from first economics course. 225 00:12:10,020 --> 00:12:12,570 It's the ratio of percentage changes. 226 00:12:12,570 --> 00:12:16,990 Limiting ratio is the changes go to 0, 227 00:12:16,990 --> 00:12:23,290 the percentage change in demand induced by a 1% change in cost, 228 00:12:23,290 --> 00:12:27,400 or an epsilon change in cost, the ratio. 229 00:12:27,400 --> 00:12:30,610 We use elasticities a lot in economics just as a reminder 230 00:12:30,610 --> 00:12:33,350 because they are unit free. 231 00:12:33,350 --> 00:12:36,080 There's no natural units here. 232 00:12:36,080 --> 00:12:39,800 So you could measure price in pennies or dollars or euros 233 00:12:39,800 --> 00:12:41,580 or whatever else. 234 00:12:41,580 --> 00:12:47,150 So you don't want to say the notion of change per dollar 235 00:12:47,150 --> 00:12:50,150 doesn't have any particular intrinsic meaning. 236 00:12:50,150 --> 00:12:55,430 The notion of percentage change from a 1% increase in cost 237 00:12:55,430 --> 00:12:57,720 is usually a little more convenient. 238 00:12:57,720 --> 00:13:01,970 So let me suppose that's my demand curve for lighting. 239 00:13:01,970 --> 00:13:04,410 And let me suppose it's the short run. 240 00:13:04,410 --> 00:13:11,570 So again, the more electricity I use, the more lighting I get. 241 00:13:11,570 --> 00:13:14,363 And e is just some constant. 242 00:13:14,363 --> 00:13:15,780 So now what I want to know, I want 243 00:13:15,780 --> 00:13:20,890 to get my demand for electricity for lighting, right? 244 00:13:20,890 --> 00:13:22,330 It's a derived demand. 245 00:13:22,330 --> 00:13:24,100 I don't care about electricity. 246 00:13:24,100 --> 00:13:29,230 I care about lighting, about L. I use electricity 247 00:13:29,230 --> 00:13:31,000 to produce it. 248 00:13:31,000 --> 00:13:34,360 The demand for lighting is derived from the demand 249 00:13:34,360 --> 00:13:37,360 for electricity. 250 00:13:37,360 --> 00:13:39,890 So how do I do it? 251 00:13:39,890 --> 00:13:46,850 OK, in this simple case, if I've got a price of electricity, 252 00:13:46,850 --> 00:13:51,740 say dollars per kilowatt hour, I could easily 253 00:13:51,740 --> 00:13:58,440 solve with this thing for a price per lumen. 254 00:13:58,440 --> 00:14:00,050 I know what electricity costs. 255 00:14:00,050 --> 00:14:03,050 I know the relation between electricity and lumens, 256 00:14:03,050 --> 00:14:03,950 it's linear. 257 00:14:03,950 --> 00:14:08,630 So I know the price, lumen hours, but lumens. 258 00:14:08,630 --> 00:14:14,220 So just to get the unit straight, 259 00:14:14,220 --> 00:14:17,010 the units of this little constant e 260 00:14:17,010 --> 00:14:22,720 has to be lumen hours per kilowatt hour, right? 261 00:14:22,720 --> 00:14:26,150 And I want a price per lumen. 262 00:14:26,150 --> 00:14:30,500 So I get $1 per lumen price by dividing the per kilowatt hour 263 00:14:30,500 --> 00:14:33,580 price by lumens per kilowatt hour. 264 00:14:33,580 --> 00:14:40,710 So I've now got the cost per lumen hour, OK? 265 00:14:40,710 --> 00:14:42,660 With me so far? 266 00:14:42,660 --> 00:14:44,802 Scream and shout if not. 267 00:14:44,802 --> 00:14:48,380 Well, don't scream, but a shout would be good. 268 00:14:48,380 --> 00:14:49,890 I can substitute then. 269 00:14:49,890 --> 00:14:57,000 I know that the amount of lumens I get 270 00:14:57,000 --> 00:15:00,780 is this little constant e times the amount of electricity. 271 00:15:03,480 --> 00:15:11,130 The demand curve then is A times the dollars 272 00:15:11,130 --> 00:15:13,620 per lumen to the minus B. That's dollars 273 00:15:13,620 --> 00:15:17,490 per lumen, PE over little e. 274 00:15:17,490 --> 00:15:20,730 Solve for electricity, and I have the demand for electricity 275 00:15:20,730 --> 00:15:25,585 as a function of its price, OK? 276 00:15:28,690 --> 00:15:31,430 Walk one more time. 277 00:15:31,430 --> 00:15:35,620 I care about lighting as a function 278 00:15:35,620 --> 00:15:38,600 of the cost of lighting. 279 00:15:38,600 --> 00:15:40,460 I don't care about electricity. 280 00:15:40,460 --> 00:15:42,980 I care about lighting as a function of its cost. 281 00:15:45,940 --> 00:15:51,720 What I want to do first is get the cost. 282 00:15:51,720 --> 00:15:56,400 And in this case, because it's the short run and very simple, 283 00:15:56,400 --> 00:15:58,110 all the bulbs are fixed. 284 00:15:58,110 --> 00:16:02,280 It's just proportional to how much electricity I use. 285 00:16:02,280 --> 00:16:08,630 So the cost-- you can think of e as relating to efficiency. 286 00:16:08,630 --> 00:16:11,130 You increase little e, basically, 287 00:16:11,130 --> 00:16:16,110 I've got more efficient bulbs, which 288 00:16:16,110 --> 00:16:17,790 is another way to get quickly. 289 00:16:17,790 --> 00:16:21,130 I do it with units because otherwise I get backwards. 290 00:16:21,130 --> 00:16:28,540 But the larger is this efficiency parameter little 291 00:16:28,540 --> 00:16:34,350 e given the electricity price, the cheaper is lighting, right? 292 00:16:34,350 --> 00:16:37,890 So it's another way to think about it. 293 00:16:37,890 --> 00:16:42,610 Once I have the cost of lighting, 294 00:16:42,610 --> 00:16:46,280 I can substitute into the demand function. 295 00:16:46,280 --> 00:16:50,570 Note that here is the production function. 296 00:16:50,570 --> 00:16:55,740 Take this second to quality and solve. 297 00:16:55,740 --> 00:16:57,690 And I get the demand for electricity 298 00:16:57,690 --> 00:17:00,060 is a function of the price of electricity 299 00:17:00,060 --> 00:17:01,770 and the efficiency of lighting. 300 00:17:06,819 --> 00:17:13,990 Question, is this easy? 301 00:17:13,990 --> 00:17:14,589 Hard? 302 00:17:14,589 --> 00:17:15,339 Confusing? 303 00:17:15,339 --> 00:17:16,130 Backwards? 304 00:17:16,130 --> 00:17:18,880 Upside down? 305 00:17:18,880 --> 00:17:19,900 Will it be on the quiz? 306 00:17:19,900 --> 00:17:20,589 Yeah, probably. 307 00:17:20,589 --> 00:17:21,790 Yeah, something like it. 308 00:17:21,790 --> 00:17:22,290 Yeah. 309 00:17:26,470 --> 00:17:28,757 Julian, Julian, you had a question? 310 00:17:28,757 --> 00:17:30,715 AUDIENCE: Oh, no, I was giving you a thumbs up. 311 00:17:30,715 --> 00:17:33,280 RICHARD SCHMALENSEE: You give me a thumbs up, OK. 312 00:17:33,280 --> 00:17:35,020 This is a rare and wonderful experience. 313 00:17:35,020 --> 00:17:38,170 Let me just treasure this for a moment. 314 00:17:38,170 --> 00:17:41,900 Does anybody have an alternative reaction of any kind. 315 00:17:41,900 --> 00:17:43,660 OK, all right, I move on. 316 00:17:46,300 --> 00:17:49,210 A couple of comments about this. 317 00:17:52,292 --> 00:17:53,375 Think about that function. 318 00:17:58,850 --> 00:18:02,810 You know, I really should have put it on this slide. 319 00:18:02,810 --> 00:18:06,260 You notice a couple of things. 320 00:18:06,260 --> 00:18:09,230 You'll notice that the elasticity of demand 321 00:18:09,230 --> 00:18:12,950 for electricity turns out to be the same as the elasticity 322 00:18:12,950 --> 00:18:15,040 of demand for lighting. 323 00:18:15,040 --> 00:18:18,210 That is not a general property. 324 00:18:18,210 --> 00:18:19,560 It can be greater or less. 325 00:18:19,560 --> 00:18:20,790 It's not a general property. 326 00:18:23,330 --> 00:18:31,110 You will also notice that what happens 327 00:18:31,110 --> 00:18:33,960 to the demand for electricity when efficiency increases 328 00:18:33,960 --> 00:18:45,210 depend on that parameter B. So that if B is greater than 1, 329 00:18:45,210 --> 00:18:49,110 then making my light bulbs more efficient 330 00:18:49,110 --> 00:18:54,680 will increase my use of electricity, right? 331 00:18:54,680 --> 00:18:57,950 Because B greater than 1 means my demand for lighting 332 00:18:57,950 --> 00:19:01,210 is really price sensitive. 333 00:19:01,210 --> 00:19:06,340 And if you've made lights more efficient, 334 00:19:06,340 --> 00:19:09,640 you've reduced the cost of lighting. 335 00:19:09,640 --> 00:19:12,130 And since my demand for lighting is price sensitive, 336 00:19:12,130 --> 00:19:13,450 you make lights more efficient. 337 00:19:13,450 --> 00:19:16,650 I use more lighting, OK. 338 00:19:20,920 --> 00:19:24,170 Those are basically the two comments. 339 00:19:24,170 --> 00:19:28,030 The first is that relationship, it's 340 00:19:28,030 --> 00:19:31,770 kind of a function of this example. 341 00:19:31,770 --> 00:19:37,220 The second comes up in debates about auto efficiency 342 00:19:37,220 --> 00:19:39,590 standards. 343 00:19:39,590 --> 00:19:42,200 Because it makes sense, if I impose 344 00:19:42,200 --> 00:19:45,760 a miles per gallon standard on cars, 345 00:19:45,760 --> 00:19:51,040 I have made driving less expensive. 346 00:19:51,040 --> 00:19:56,560 And if driving is really price sensitive, 347 00:19:56,560 --> 00:19:59,815 that could on balance increase the demand for gasoline. 348 00:20:02,610 --> 00:20:05,040 Now, this gets debated. 349 00:20:05,040 --> 00:20:07,070 This is called the rebound effect 350 00:20:07,070 --> 00:20:12,420 in talking about CAFE standards, auto fuel efficiency standards. 351 00:20:12,420 --> 00:20:19,380 Most estimates say that it happens. 352 00:20:19,380 --> 00:20:23,750 But as far as CAFE standards are concerned, in effect, 353 00:20:23,750 --> 00:20:24,653 B is less than 1. 354 00:20:24,653 --> 00:20:25,820 The rebound effect is small. 355 00:20:25,820 --> 00:20:27,000 But it's there. 356 00:20:27,000 --> 00:20:30,260 Estimates show it to be there because driving 357 00:20:30,260 --> 00:20:32,180 is price sensitive. 358 00:20:32,180 --> 00:20:34,340 Price of gasoline goes up, people drive less. 359 00:20:34,340 --> 00:20:36,830 Cars get cheaper to drive, people drive more. 360 00:20:36,830 --> 00:20:39,680 Not a big effect, not a huge effect, but it's there. 361 00:20:42,330 --> 00:20:43,290 OK so far? 362 00:20:47,560 --> 00:20:48,970 Very quiet today, OK. 363 00:20:52,090 --> 00:20:55,010 All the enthusiasm went into negotiations then, OK. 364 00:20:58,060 --> 00:21:01,450 Obviously price isn't the only thing that matters. 365 00:21:01,450 --> 00:21:04,530 I find a very interesting graph. 366 00:21:04,530 --> 00:21:08,310 This is passenger kilometers traveled per capita 367 00:21:08,310 --> 00:21:09,570 on the vertical axis. 368 00:21:09,570 --> 00:21:12,150 It's a log, log scale. 369 00:21:12,150 --> 00:21:15,930 And GDP per capita on the horizontal 370 00:21:15,930 --> 00:21:19,320 axis, you will notice that the relation looks linear. 371 00:21:19,320 --> 00:21:22,140 What you can't see is that this is 372 00:21:22,140 --> 00:21:25,710 a scatter that includes all kinds of different countries, 373 00:21:25,710 --> 00:21:34,380 sub-Saharan Africa, Latin America, North America. 374 00:21:34,380 --> 00:21:38,140 The universe is somewhere in this scatter over time. 375 00:21:38,140 --> 00:21:41,730 So this is probably North America. 376 00:21:41,730 --> 00:21:44,790 It points to a relatively constant elasticity. 377 00:21:44,790 --> 00:21:48,040 That's not driven, that's traveled. 378 00:21:48,040 --> 00:21:53,680 So you get airplanes at the top, and so forth. 379 00:21:53,680 --> 00:21:56,440 There is this law, and I don't remember the parameter that 380 00:21:56,440 --> 00:21:59,920 says you spend x% your time traveling, 381 00:21:59,920 --> 00:22:03,850 whether you're a poor person walking or a rich person 382 00:22:03,850 --> 00:22:06,340 flying, the travel percentage turns out 383 00:22:06,340 --> 00:22:08,560 to not vary much with income. 384 00:22:08,560 --> 00:22:10,565 That's a big income effect. 385 00:22:10,565 --> 00:22:11,440 Let me just say that. 386 00:22:11,440 --> 00:22:14,530 That's an income elasticity floating around 1, 387 00:22:14,530 --> 00:22:17,110 double income, double vehicle miles traveled. 388 00:22:17,110 --> 00:22:19,660 It doesn't necessarily mean double gasoline, 389 00:22:19,660 --> 00:22:22,030 doubling anything else, people travel other ways. 390 00:22:22,030 --> 00:22:27,760 But income affects energy demand, obviously, a lot. 391 00:22:27,760 --> 00:22:29,260 What about the long run? 392 00:22:33,220 --> 00:22:37,960 Well, when I talked about the demand for lighting 393 00:22:37,960 --> 00:22:43,770 or the demand for driving or anything else, 394 00:22:43,770 --> 00:22:49,120 in that simple example, or in the example in the note, 395 00:22:49,120 --> 00:22:52,330 you hold the capital stock fixed. 396 00:22:52,330 --> 00:22:54,430 In the longer run, and here gasoline 397 00:22:54,430 --> 00:23:00,770 is the thing to think about, I'll change my car. 398 00:23:00,770 --> 00:23:04,163 So in the short run, price of gasoline moves. 399 00:23:04,163 --> 00:23:05,330 Maybe I drive a little more. 400 00:23:05,330 --> 00:23:07,810 Maybe I drive a little less. 401 00:23:07,810 --> 00:23:13,490 Price of gasoline goes up by a multiple of 5 and stays there, 402 00:23:13,490 --> 00:23:15,960 let's say. 403 00:23:15,960 --> 00:23:18,870 I probably would change my car. 404 00:23:18,870 --> 00:23:22,080 I might want to consider changing where I live. 405 00:23:22,080 --> 00:23:24,900 I might want to rethink my use of public transportation. 406 00:23:24,900 --> 00:23:27,720 I can do a lot of things in the longer run. 407 00:23:27,720 --> 00:23:31,140 I can certainly change my car and think about where I live. 408 00:23:31,140 --> 00:23:35,220 Do I really want to live in Weston as opposed to-- oh, 409 00:23:35,220 --> 00:23:36,120 you're not from here. 410 00:23:36,120 --> 00:23:39,810 Do I really want to live way out in the suburbs 411 00:23:39,810 --> 00:23:41,490 if gasoline is really expensive. 412 00:23:41,490 --> 00:23:45,820 Or do I want to think about moving into town. 413 00:23:45,820 --> 00:23:49,660 And if I'm locating a factory, and I'm thinking about, 414 00:23:49,660 --> 00:23:55,390 do workers like to drive very far, or an office building, 415 00:23:55,390 --> 00:23:57,500 I have to rethink where to put it. 416 00:23:57,500 --> 00:24:02,740 So in the longer run, you expect more change. 417 00:24:02,740 --> 00:24:06,160 You expect whatever the response to an energy price 418 00:24:06,160 --> 00:24:09,640 changes in the short run, bip, bip, 419 00:24:09,640 --> 00:24:12,820 bip, you expect if it's maintained, 420 00:24:12,820 --> 00:24:14,380 the response will be longer. 421 00:24:14,380 --> 00:24:15,880 That will be larger. 422 00:24:15,880 --> 00:24:19,040 And I will come back to this. 423 00:24:19,040 --> 00:24:20,830 This is sort of a basic principle. 424 00:24:20,830 --> 00:24:24,280 Paul Samuelson was the first to really lay it out. 425 00:24:24,280 --> 00:24:27,310 You have more flexibility in the long run, more things, 426 00:24:27,310 --> 00:24:29,380 more degrees of freedom. 427 00:24:29,380 --> 00:24:31,960 You expect a greater response to almost anything, 428 00:24:31,960 --> 00:24:33,880 income change, price change, whatever, 429 00:24:33,880 --> 00:24:35,980 if it's maintained over time. 430 00:24:35,980 --> 00:24:38,260 You expect the response to grow to a change that's 431 00:24:38,260 --> 00:24:39,370 maintained over time. 432 00:24:42,240 --> 00:24:45,990 Part of that is because of the long-lived capital. 433 00:24:49,120 --> 00:24:53,410 But you will also recall that we talked about all kinds 434 00:24:53,410 --> 00:24:55,670 of path-dependent reasons. 435 00:24:55,670 --> 00:24:58,510 So let me just slip outside economics for just 436 00:24:58,510 --> 00:25:04,050 a slight moment and say that for the pure economic side, one 437 00:25:04,050 --> 00:25:07,350 reason why the long run response is greater 438 00:25:07,350 --> 00:25:10,200 than the short run response is because you change capital 439 00:25:10,200 --> 00:25:11,040 stock. 440 00:25:11,040 --> 00:25:15,210 And you don't change capital stock quickly. 441 00:25:15,210 --> 00:25:17,850 At the societal level, the things 442 00:25:17,850 --> 00:25:22,740 that affect response may be policies, may be habits, 443 00:25:22,740 --> 00:25:24,540 may be culture. 444 00:25:24,540 --> 00:25:27,390 And they change very quickly. 445 00:25:27,390 --> 00:25:29,580 For Boston to look like Copenhagen 446 00:25:29,580 --> 00:25:32,630 as far as bicycle usage is concerned 447 00:25:32,630 --> 00:25:36,390 would not be easy or quick. 448 00:25:36,390 --> 00:25:38,160 They're having ferocious debates about how 449 00:25:38,160 --> 00:25:42,330 to set up the Longfellow bridge for bicycles versus cars. 450 00:25:42,330 --> 00:25:47,500 And that bridge hasn't been reconfigured in 100 years. 451 00:25:47,500 --> 00:25:51,620 It will now have, I think, more bicycle space going forward. 452 00:25:51,620 --> 00:25:53,253 And if you've biked across that bridge, 453 00:25:53,253 --> 00:25:54,670 you will think that's a good thing 454 00:25:54,670 --> 00:25:56,660 because it's a little tricky-- 455 00:25:56,660 --> 00:25:59,720 so long run, short run. 456 00:25:59,720 --> 00:26:04,340 So let me now talk about how you estimate these things. 457 00:26:04,340 --> 00:26:05,420 So that's theory. 458 00:26:08,870 --> 00:26:11,300 I'm a rational calculator, I'm Walmart, 459 00:26:11,300 --> 00:26:14,360 and I want to know my demand for electricity 460 00:26:14,360 --> 00:26:15,620 for heating and cooling. 461 00:26:15,620 --> 00:26:16,975 I don't care about electricity. 462 00:26:16,975 --> 00:26:18,350 I care about heating and cooling. 463 00:26:18,350 --> 00:26:21,500 I want to think through how my demand for electricity 464 00:26:21,500 --> 00:26:25,490 will be reflected by changes in the price. 465 00:26:25,490 --> 00:26:27,410 And I will look at my production function, 466 00:26:27,410 --> 00:26:29,540 the HVAC system I've got. 467 00:26:29,540 --> 00:26:31,610 And I'll look at my demand function, 468 00:26:31,610 --> 00:26:34,160 how much do I care about store temperature 469 00:26:34,160 --> 00:26:39,470 and figure out what I want to do in terms of electricity. 470 00:26:39,470 --> 00:26:41,150 And then in the longer run, depending 471 00:26:41,150 --> 00:26:44,240 on the cost of electricity, I will insulate more or less. 472 00:26:44,240 --> 00:26:46,450 I will design stores in different ways. 473 00:26:46,450 --> 00:26:49,560 I have all kinds of response options. 474 00:26:49,560 --> 00:26:53,720 So that's terrific for the decision maker. 475 00:26:53,720 --> 00:26:59,042 Suppose you want to estimate the demand for electricity, 476 00:26:59,042 --> 00:27:00,500 the demand for gasoline or anything 477 00:27:00,500 --> 00:27:03,260 else by looking at the market. 478 00:27:03,260 --> 00:27:04,280 You're a seller. 479 00:27:04,280 --> 00:27:05,540 You're a policymaker. 480 00:27:05,540 --> 00:27:07,610 You care about what all of that stuff 481 00:27:07,610 --> 00:27:10,880 ends up looking like at the end of the day when 482 00:27:10,880 --> 00:27:12,890 all the decisions shake out. 483 00:27:12,890 --> 00:27:18,610 So a natural thing to do is to look at data. 484 00:27:18,610 --> 00:27:23,888 The first place this was done was in agriculture. 485 00:27:23,888 --> 00:27:25,430 People said well, you know, we'd like 486 00:27:25,430 --> 00:27:27,200 to know what the demand curve for wheat looks like, 487 00:27:27,200 --> 00:27:28,700 what the demand curve for corn looks 488 00:27:28,700 --> 00:27:31,220 like for planning purposes. 489 00:27:31,220 --> 00:27:33,140 If you've seen the Department of Agriculture 490 00:27:33,140 --> 00:27:35,240 building in Washington, it was for a long time 491 00:27:35,240 --> 00:27:38,060 the largest building in the world. 492 00:27:38,060 --> 00:27:42,180 We spend a lot of effort studying agricultural markets. 493 00:27:42,180 --> 00:27:44,840 And when they first did this they 494 00:27:44,840 --> 00:27:47,900 found demand curves that sloped up. 495 00:27:47,900 --> 00:27:50,030 They graphed price, and they graphed quantity, 496 00:27:50,030 --> 00:27:52,890 and they got demand curves sloping up. 497 00:27:52,890 --> 00:27:53,970 This was disturbing. 498 00:27:53,970 --> 00:27:54,920 How could that happen. 499 00:28:01,450 --> 00:28:02,937 Yes, Marie. 500 00:28:02,937 --> 00:28:04,520 AUDIENCE: Well, the demand curve could 501 00:28:04,520 --> 00:28:05,930 shift with the supply curve. 502 00:28:05,930 --> 00:28:09,068 And if they shift more easily, you might a bad-- 503 00:28:09,068 --> 00:28:11,485 your data is not going to follow the trend of specifically 504 00:28:11,485 --> 00:28:13,660 the demand curve. 505 00:28:13,660 --> 00:28:15,490 RICHARD SCHMALENSEE: OK, close. 506 00:28:15,490 --> 00:28:17,560 You're saying that the data are going 507 00:28:17,560 --> 00:28:20,140 to be affected by the supply curve as well as the demand 508 00:28:20,140 --> 00:28:21,170 curve. 509 00:28:21,170 --> 00:28:24,777 In fact, suppose the supply curve doesn't shift at all. 510 00:28:24,777 --> 00:28:25,860 What are you going to get? 511 00:28:28,208 --> 00:28:30,500 AUDIENCE: If the supply curve doesn't shift at all then 512 00:28:30,500 --> 00:28:33,370 you're going to get every change in-- 513 00:28:33,370 --> 00:28:35,470 like the equilibrium point is going 514 00:28:35,470 --> 00:28:38,145 to be a point along the demand curve. 515 00:28:38,145 --> 00:28:39,520 RICHARD SCHMALENSEE: Think again. 516 00:28:39,520 --> 00:28:42,320 The supply curve doesn't shift, the demand curve does shift. 517 00:28:42,320 --> 00:28:43,240 What do you get? 518 00:28:43,240 --> 00:28:44,470 AUDIENCE: You are going to trace out the supply curve. 519 00:28:44,470 --> 00:28:44,590 RICHARD SCHMALENSEE: You're going 520 00:28:44,590 --> 00:28:46,423 to trace out the supply curve, which is what 521 00:28:46,423 --> 00:28:47,740 David was dying to say, right? 522 00:28:47,740 --> 00:28:49,240 Am I right? 523 00:28:49,240 --> 00:28:51,190 OK. 524 00:28:51,190 --> 00:28:54,550 We're going to move to thumbs, OK? 525 00:28:54,550 --> 00:28:59,760 The classic picture, the classic picture, so you have what's 526 00:28:59,760 --> 00:29:01,770 come to be called an Identification problem. 527 00:29:01,770 --> 00:29:07,940 How do you figure out, how do you get to demand? 528 00:29:07,940 --> 00:29:10,730 I care about demand. 529 00:29:10,730 --> 00:29:11,783 I'm not getting demand. 530 00:29:11,783 --> 00:29:12,575 I'm getting supply. 531 00:29:15,630 --> 00:29:18,300 And in fact, if that's the world, 532 00:29:18,300 --> 00:29:20,460 if all I know were price and quantity, 533 00:29:20,460 --> 00:29:22,530 and all I know is the demand curve shifts 534 00:29:22,530 --> 00:29:26,520 and the supply curve doesn't, I can't estimate demand. 535 00:29:26,520 --> 00:29:28,870 That's to say, it's not identified. 536 00:29:28,870 --> 00:29:32,820 I could gather all the data in the world from that setting. 537 00:29:32,820 --> 00:29:37,020 And I could not learn about demand from the data. 538 00:29:37,020 --> 00:29:39,580 Demand curve is not identified. 539 00:29:39,580 --> 00:29:42,385 So what you need-- 540 00:29:45,930 --> 00:29:47,520 and Marie was on to it. 541 00:29:47,520 --> 00:29:52,250 What you need is you've got to move the supply curve. 542 00:29:52,250 --> 00:29:55,257 There has to be something that affects supply. 543 00:29:55,257 --> 00:29:57,090 If you've got something that affects supply, 544 00:29:57,090 --> 00:29:59,298 and you've got something that affects demand and they 545 00:29:59,298 --> 00:30:01,142 both move around, then if you work 546 00:30:01,142 --> 00:30:02,850 through the example in the teaching note, 547 00:30:02,850 --> 00:30:05,970 you can, in principle estimate both of them. 548 00:30:08,780 --> 00:30:10,560 There are nice techniques. 549 00:30:10,560 --> 00:30:11,780 This is a lot of fun. 550 00:30:11,780 --> 00:30:15,290 But I mention this not because this course 551 00:30:15,290 --> 00:30:19,010 is about learning these techniques, which it's not, 552 00:30:19,010 --> 00:30:23,960 but it is about being skeptical when people tell you, well, we 553 00:30:23,960 --> 00:30:25,580 plotted price and quantity and we 554 00:30:25,580 --> 00:30:27,910 got-- here's our demand curve. 555 00:30:27,910 --> 00:30:33,910 Whenever you're dealing with a demand supply situation, 556 00:30:33,910 --> 00:30:36,850 you have to ask, how do you know that you're actually 557 00:30:36,850 --> 00:30:40,640 getting demand not supply not some mixture? 558 00:30:40,640 --> 00:30:41,140 Right? 559 00:30:41,140 --> 00:30:43,690 You can have both demand and supply shifting, 560 00:30:43,690 --> 00:30:46,560 and you could get a scatter of points 561 00:30:46,560 --> 00:30:50,260 that slopes down or is flat. 562 00:30:50,260 --> 00:30:52,720 You could say, well, you know, demand is very price elastic 563 00:30:52,720 --> 00:30:54,130 because the scatter shows this. 564 00:30:54,130 --> 00:30:56,910 Well, it doesn't. 565 00:30:56,910 --> 00:31:00,060 This is just a caution. 566 00:31:00,060 --> 00:31:03,210 Because you will see in your life, 567 00:31:03,210 --> 00:31:05,310 I expect, with reasonable probability, 568 00:31:05,310 --> 00:31:08,680 estimates of the demand for this and the demand for that. 569 00:31:08,680 --> 00:31:12,270 And question one, is how do you know you've got demand? 570 00:31:12,270 --> 00:31:15,978 What did you do to deal with this identification problem? 571 00:31:15,978 --> 00:31:17,520 What did you do to deal with the fact 572 00:31:17,520 --> 00:31:22,380 that demand and supply are simultaneously determined? 573 00:31:22,380 --> 00:31:33,520 OK, clear, simple, classic stuff, OK. 574 00:31:33,520 --> 00:31:44,920 Now, I talked about the long run and the short run. 575 00:31:44,920 --> 00:31:47,610 So how do we do that in practice? 576 00:31:47,610 --> 00:31:56,310 Well, there's a simple way that's pretty commonly used. 577 00:31:56,310 --> 00:32:00,040 And it's called a partial adjustment model. 578 00:32:00,040 --> 00:32:04,440 And the equation let's say, E is the demand for energy. 579 00:32:04,440 --> 00:32:06,190 And here all the things that determine it. 580 00:32:06,190 --> 00:32:08,320 E star is a function. 581 00:32:08,320 --> 00:32:12,040 E star is considered to be the long run demand function. 582 00:32:12,040 --> 00:32:16,910 And lambda is a constant between 0 and 1, OK? 583 00:32:16,910 --> 00:32:18,800 So this is a common formulation. 584 00:32:18,800 --> 00:32:23,780 If you think about it and you fiddled with difference 585 00:32:23,780 --> 00:32:27,680 equations, or you can just substitute, 586 00:32:27,680 --> 00:32:29,600 suppose E star is constant. 587 00:32:29,600 --> 00:32:33,170 Suppose the demand for energy is constant 588 00:32:33,170 --> 00:32:35,480 because the prices and other things have changed, 589 00:32:35,480 --> 00:32:38,790 and they've changed permanently. 590 00:32:38,790 --> 00:32:44,260 Then E is going to converge to E star. 591 00:32:44,260 --> 00:32:50,020 Again, you just start out at some E0 and just substitute. 592 00:32:50,020 --> 00:32:53,380 And you will see that if lambda is less than 1, 593 00:32:53,380 --> 00:32:58,560 the initial condition goes away and it converges to E star. 594 00:32:58,560 --> 00:33:00,520 So this is a model. 595 00:33:00,520 --> 00:33:09,330 But in the short run, in the short run, the change in demand 596 00:33:09,330 --> 00:33:13,670 will be reduced by this multiple lambda, right? 597 00:33:13,670 --> 00:33:21,850 So suppose doubling income doubles E star. 598 00:33:21,850 --> 00:33:24,790 OK, that's the long run response. 599 00:33:24,790 --> 00:33:33,630 If lambda is 0.5, then doubling income today 600 00:33:33,630 --> 00:33:42,010 will by itself only increase energy by 50% of E star, right? 601 00:33:42,010 --> 00:33:44,520 So it's a smoothing. 602 00:33:44,520 --> 00:33:46,890 You could think about this as smoothing as signal 603 00:33:46,890 --> 00:33:48,750 detection is whatever you like. 604 00:33:48,750 --> 00:33:51,240 But the effective lambda less than 1 605 00:33:51,240 --> 00:33:53,040 is that the response in the short run 606 00:33:53,040 --> 00:33:54,870 is less than the response in the long run. 607 00:33:58,610 --> 00:34:00,265 So you estimate that. 608 00:34:03,420 --> 00:34:05,860 You assume a function that involves everything 609 00:34:05,860 --> 00:34:07,330 else you can think of. 610 00:34:07,330 --> 00:34:09,850 You use some technique to deal with the fact 611 00:34:09,850 --> 00:34:14,110 that price is determined along with quantity, along with E 612 00:34:14,110 --> 00:34:19,190 and estimate a function of that form, pretty commonly done, 613 00:34:19,190 --> 00:34:22,100 pretty commonly done. 614 00:34:22,100 --> 00:34:24,830 I'm going to give you some actual estimates in a minute. 615 00:34:24,830 --> 00:34:29,489 They tend to come from a framework like that where you 616 00:34:29,489 --> 00:34:33,840 say, it's a simple way of saying if a change is maintained 617 00:34:33,840 --> 00:34:35,969 over time, the response will be greater 618 00:34:35,969 --> 00:34:40,670 than if the change happens today and gets reversed tomorrow. 619 00:34:40,670 --> 00:34:46,989 It has that property for lambda between 0 and 1, OK? 620 00:34:50,810 --> 00:34:53,940 Yeah, that's all straightforward. 621 00:34:53,940 --> 00:34:56,380 You can recover so to speak. 622 00:34:59,150 --> 00:35:01,520 You estimate a function like this 623 00:35:01,520 --> 00:35:04,260 that has lagged demand on the right as well. 624 00:35:07,040 --> 00:35:10,190 Its coefficient is an estimate of 1 minus lambda. 625 00:35:10,190 --> 00:35:11,840 That lets you recover lambda. 626 00:35:11,840 --> 00:35:15,860 That lets you get the long run function, right? 627 00:35:15,860 --> 00:35:18,430 So if I want to know what the long run response is, 628 00:35:18,430 --> 00:35:22,550 I look at the lag dependent variable. 629 00:35:22,550 --> 00:35:26,890 And again, if you think about it, 630 00:35:26,890 --> 00:35:30,290 suppose this coefficient is 0.9. 631 00:35:30,290 --> 00:35:32,690 That means all else equal, E is going 632 00:35:32,690 --> 00:35:36,140 to tend to be pretty smooth, right? 633 00:35:36,140 --> 00:35:39,470 The changes in E star are going to be highly damp. 634 00:35:39,470 --> 00:35:40,910 This will be 0.1. 635 00:35:40,910 --> 00:35:43,055 And E itself is going to be pretty smooth. 636 00:35:45,710 --> 00:35:48,770 But I can sort of undo that by saying, all right, 637 00:35:48,770 --> 00:35:52,670 I'll divide by 0.1 over here and get 638 00:35:52,670 --> 00:35:57,560 an estimate of the function that would govern response 639 00:35:57,560 --> 00:36:03,870 to permanent changes, OK? 640 00:36:03,870 --> 00:36:05,430 There are other approaches. 641 00:36:05,430 --> 00:36:10,290 There's a note, the note cites this very interesting paper 642 00:36:10,290 --> 00:36:14,640 by Hill Huntington that looks at oil price changes. 643 00:36:14,640 --> 00:36:17,490 He notes that the price of oil at any time 644 00:36:17,490 --> 00:36:21,630 is equal to the sum of the maximum it's ever been. 645 00:36:21,630 --> 00:36:23,740 And the difference between what it is today 646 00:36:23,740 --> 00:36:25,410 and the maximum it's ever been. 647 00:36:25,410 --> 00:36:27,960 He puts both of these into an equation 648 00:36:27,960 --> 00:36:32,160 and finds that the maximum has the largest negative effect 649 00:36:32,160 --> 00:36:35,430 on demand for gasoline. 650 00:36:35,430 --> 00:36:38,430 In his argument-- so it's not this. 651 00:36:38,430 --> 00:36:40,560 It's a different adjustment mechanism. 652 00:36:40,560 --> 00:36:44,718 He says people get a sense of what kind of cars 653 00:36:44,718 --> 00:36:46,260 they ought to have and how they ought 654 00:36:46,260 --> 00:36:49,950 to heat their houses based on how high oil or gasoline has 655 00:36:49,950 --> 00:36:51,890 been. 656 00:36:51,890 --> 00:36:56,780 So the previous maximum has a very powerful shock impact. 657 00:36:56,780 --> 00:37:00,200 We do remember gasoline prices over $4 a gallon. 658 00:37:00,200 --> 00:37:04,430 You may be able to say what it is right now, I can't. 659 00:37:04,430 --> 00:37:09,320 So P max the argument is has a large effect. 660 00:37:09,320 --> 00:37:10,670 I'm not sure that's right. 661 00:37:10,670 --> 00:37:13,460 I haven't seen any other study that tested that effectively 662 00:37:13,460 --> 00:37:14,990 or that reproduced it. 663 00:37:14,990 --> 00:37:19,010 But it it's an interesting approach. 664 00:37:19,010 --> 00:37:22,810 Most of the time, again, the intuition, 665 00:37:22,810 --> 00:37:25,030 because all this stuff is durable, 666 00:37:25,030 --> 00:37:32,005 all these assets are durable, changes take a long time, OK. 667 00:37:34,760 --> 00:37:40,710 Some results, this is something that people 668 00:37:40,710 --> 00:37:46,800 have studied a lot, the demand for various forms of energy. 669 00:37:46,800 --> 00:37:51,530 And I will have assigned a paper or two. 670 00:37:51,530 --> 00:37:55,790 But they all go properly into the deep weeds, 671 00:37:55,790 --> 00:37:57,840 technically, and in terms of data. 672 00:37:57,840 --> 00:38:01,100 So I'm just going to give you a summary. 673 00:38:01,100 --> 00:38:04,010 The first summary point is the difference 674 00:38:04,010 --> 00:38:07,410 is normally huge between the short run and the long run. 675 00:38:07,410 --> 00:38:10,220 So to go back to that other formulation, 676 00:38:10,220 --> 00:38:14,240 it's as if lambda is generally quite 677 00:38:14,240 --> 00:38:19,390 small in models like that, estimated to be quite small. 678 00:38:23,630 --> 00:38:29,280 Second, short run income and price elasticity, say a year, 679 00:38:29,280 --> 00:38:31,950 is generally less than 0.5. 680 00:38:31,950 --> 00:38:37,340 So if you double price, you will reduce consumption 681 00:38:37,340 --> 00:38:39,680 by less than half. 682 00:38:39,680 --> 00:38:44,180 I mean generally, you will see a good deal less, 683 00:38:44,180 --> 00:38:46,730 double income price consumption will 684 00:38:46,730 --> 00:38:48,800 increase by less than half. 685 00:38:54,450 --> 00:38:56,970 The results seem to depend on the economy, 686 00:38:56,970 --> 00:38:59,070 rich countries, poor countries. 687 00:38:59,070 --> 00:39:01,290 That might make some sense, density. 688 00:39:01,290 --> 00:39:02,790 You can think of all kinds of things 689 00:39:02,790 --> 00:39:04,860 that might affect how people respond 690 00:39:04,860 --> 00:39:07,050 to changes in price and income. 691 00:39:07,050 --> 00:39:13,170 A lot of the studies have been of gasoline and electricity. 692 00:39:13,170 --> 00:39:17,010 And here's some of the ranges for elasticities. 693 00:39:19,610 --> 00:39:23,120 Income elasticity is the percentage increase 694 00:39:23,120 --> 00:39:25,760 in demand by 1%-- 695 00:39:25,760 --> 00:39:29,040 small 1%-- increase in income. 696 00:39:29,040 --> 00:39:32,420 So income elasticity of 1 means if I double your income, 697 00:39:32,420 --> 00:39:35,590 you double your demand. 698 00:39:35,590 --> 00:39:38,470 The short run elasticities for gasoline-- 699 00:39:41,480 --> 00:39:44,240 yeah, the short run price elasticities 700 00:39:44,240 --> 00:39:47,510 for gasoline and electricity tend to be quite low. 701 00:39:51,920 --> 00:39:53,510 Just looking at the literature, I 702 00:39:53,510 --> 00:39:56,180 think that the studies tend to cluster more 703 00:39:56,180 --> 00:39:59,480 toward the low end of those ranges. 704 00:39:59,480 --> 00:40:04,150 So price elasticity of 0.1 means I double the price. 705 00:40:04,150 --> 00:40:07,760 Consumption goes down by 10%. 706 00:40:07,760 --> 00:40:10,720 0.2 means 20%. 707 00:40:10,720 --> 00:40:13,150 So these are small responses. 708 00:40:13,150 --> 00:40:15,090 These are low elasticities. 709 00:40:15,090 --> 00:40:18,540 Notice there's a big difference-- 710 00:40:18,540 --> 00:40:22,310 factor of 2, 2 and 1/2 there. 711 00:40:22,310 --> 00:40:26,540 Bigger in gasoline than electricity, 712 00:40:26,540 --> 00:40:30,520 which makes a little more sense but only a little. 713 00:40:33,460 --> 00:40:35,410 The income elasticity of electricity I'm 714 00:40:35,410 --> 00:40:37,180 going to return to. 715 00:40:37,180 --> 00:40:40,670 In the long run, that's tended to be pretty high. 716 00:40:40,670 --> 00:40:43,410 Most estimates say, as people get richer, 717 00:40:43,410 --> 00:40:45,560 they use more electricity. 718 00:40:45,560 --> 00:40:50,750 And it's plus or minus proportional to income. 719 00:40:50,750 --> 00:40:53,840 You add appliances, bigger house. 720 00:40:53,840 --> 00:40:55,820 All the things that take electricity 721 00:40:55,820 --> 00:40:57,290 tend to go up with income. 722 00:40:57,290 --> 00:40:59,030 And that affect is not small. 723 00:40:59,030 --> 00:41:02,580 In the short run, well, that's pretty small, right? 724 00:41:02,580 --> 00:41:03,530 I mean. 725 00:41:03,530 --> 00:41:07,910 My income goes up significantly in the short run. 726 00:41:07,910 --> 00:41:09,320 What do I use electricity for? 727 00:41:09,320 --> 00:41:11,100 It's mostly related. 728 00:41:11,100 --> 00:41:15,000 These are residential, mostly residential. 729 00:41:15,000 --> 00:41:16,710 There's relatively little I can change 730 00:41:16,710 --> 00:41:20,000 until I start changing assets. 731 00:41:20,000 --> 00:41:23,350 So it makes sense that again, there's 732 00:41:23,350 --> 00:41:25,000 a big long run, short-run gap. 733 00:41:25,000 --> 00:41:27,340 What's interesting here is the long-run elasticities 734 00:41:27,340 --> 00:41:30,490 are pretty big historically. 735 00:41:30,490 --> 00:41:32,220 Yeah, Andrew. 736 00:41:32,220 --> 00:41:35,220 AUDIENCE: Do these numbers-- 737 00:41:35,220 --> 00:41:40,455 they seem to take on something from the '70s to now. 738 00:41:40,455 --> 00:41:42,830 RICHARD SCHMALENSEE: No, I'm going to come right to that. 739 00:41:46,220 --> 00:41:49,310 The width of those ranges ought to tell you that these are not 740 00:41:49,310 --> 00:41:52,220 constants of nature. 741 00:41:52,220 --> 00:41:54,980 This is a summary of human behavior, human behavior 742 00:41:54,980 --> 00:41:56,990 changes. 743 00:41:56,990 --> 00:41:58,970 So they do change over time. 744 00:42:04,990 --> 00:42:08,480 And the problem, there is a piece 745 00:42:08,480 --> 00:42:11,967 I could have assigned because it's pretty simple. 746 00:42:11,967 --> 00:42:13,550 There's a piece that's argued, there's 747 00:42:13,550 --> 00:42:17,300 been a big change in the price elasticity of demand 748 00:42:17,300 --> 00:42:19,160 for gasoline in the '80s. 749 00:42:19,160 --> 00:42:21,650 And I'm trying to remember whether it goes up or down. 750 00:42:21,650 --> 00:42:25,580 So people have looked at this, but not much, not much. 751 00:42:25,580 --> 00:42:28,040 You just see different studies from different periods 752 00:42:28,040 --> 00:42:30,720 and in different countries getting different results. 753 00:42:30,720 --> 00:42:33,410 And you say, well, maybe that's a change in reality 754 00:42:33,410 --> 00:42:37,050 or maybe that's a different method, right? 755 00:42:37,050 --> 00:42:39,045 It depends on data. 756 00:42:39,045 --> 00:42:41,940 It depends on, can you study households? 757 00:42:41,940 --> 00:42:43,157 Is this aggregate data? 758 00:42:43,157 --> 00:42:44,240 What are you working with? 759 00:42:47,110 --> 00:42:49,720 But I want to make a couple of points 760 00:42:49,720 --> 00:42:51,730 about this stuff being constants of nature. 761 00:42:54,290 --> 00:43:00,010 This is one of the interesting puzzles. 762 00:43:00,010 --> 00:43:03,140 This is electricity demand. 763 00:43:03,140 --> 00:43:06,020 And if you look at electricity use in the United States 764 00:43:06,020 --> 00:43:12,460 before about 1970, you get sort of the top two rows. 765 00:43:12,460 --> 00:43:15,940 You get pretty good growth in GDP. 766 00:43:15,940 --> 00:43:17,540 And these are not per capita. 767 00:43:17,540 --> 00:43:22,060 So this is total electricity use, total GDP. 768 00:43:22,060 --> 00:43:26,200 That's pretty good annual growth rates, 3 and 1/2, 4, 769 00:43:26,200 --> 00:43:30,120 and very big growth rates in electricity. 770 00:43:30,120 --> 00:43:35,195 So the conventional wisdom was the income elasticity 771 00:43:35,195 --> 00:43:36,380 is greater than 1. 772 00:43:36,380 --> 00:43:39,160 And the price elasticity is small. 773 00:43:39,160 --> 00:43:44,550 Now, there really wasn't much historical experience 774 00:43:44,550 --> 00:43:48,690 you could base that on except the economy had grown. 775 00:43:48,690 --> 00:43:50,580 And if you go back farther, the figures 776 00:43:50,580 --> 00:43:53,370 get even bigger for growth in electricity. 777 00:43:53,370 --> 00:43:57,500 But I didn't have everything back far enough. 778 00:43:57,500 --> 00:44:03,140 The experience was GDP grows pretty smoothly. 779 00:44:03,140 --> 00:44:06,020 Not much falloff in electricity in recessions. 780 00:44:06,020 --> 00:44:09,290 So we're looking here at the long run over decades. 781 00:44:09,290 --> 00:44:12,980 And prices do vary from time to time 782 00:44:12,980 --> 00:44:15,680 but nobody can see an effect. 783 00:44:15,680 --> 00:44:19,560 So we come now to the '70s. 784 00:44:19,560 --> 00:44:20,850 And there's an oil shock. 785 00:44:23,800 --> 00:44:26,230 And what happens is a consequence of the oil shock 786 00:44:26,230 --> 00:44:29,290 is all kinds of inputs to electricity go up. 787 00:44:29,290 --> 00:44:32,770 And instead of declining faster than other prices, 788 00:44:32,770 --> 00:44:35,110 the price of electricity is rising faster 789 00:44:35,110 --> 00:44:37,540 than other prices. 790 00:44:37,540 --> 00:44:46,570 So what do you think happened to the total consumption 791 00:44:46,570 --> 00:44:49,210 of electricity? 792 00:44:49,210 --> 00:44:52,252 How many think it went up by 1% to 2%? 793 00:44:55,204 --> 00:44:56,195 2% to 3%? 794 00:45:01,070 --> 00:45:02,405 Thank you, Julian. 795 00:45:05,150 --> 00:45:06,110 3 to 4? 796 00:45:09,050 --> 00:45:11,360 How many think we shut off the electricity? 797 00:45:11,360 --> 00:45:14,220 4 to 5. 798 00:45:14,220 --> 00:45:15,420 Oh, come on. 799 00:45:15,420 --> 00:45:16,230 You've got to vote. 800 00:45:16,230 --> 00:45:17,280 4 to 5, all right. 801 00:45:21,080 --> 00:45:24,320 It went up by 4. 802 00:45:24,320 --> 00:45:29,710 So you immediately say, wow, income growth actually wasn't 803 00:45:29,710 --> 00:45:31,900 that bad over that period. 804 00:45:31,900 --> 00:45:35,540 There must be more price sensitivity than we thought. 805 00:45:35,540 --> 00:45:36,898 We got it wrong. 806 00:45:36,898 --> 00:45:39,440 And we were fooled by the fact that we didn't have much price 807 00:45:39,440 --> 00:45:40,220 variation. 808 00:45:40,220 --> 00:45:45,290 Prices moved and growth slowed sharply 809 00:45:45,290 --> 00:45:47,330 from those earlier numbers. 810 00:45:47,330 --> 00:45:53,830 So we come now to the decade of the '80s. 811 00:45:53,830 --> 00:45:56,630 Income growth picks up. 812 00:45:56,630 --> 00:45:58,370 Prices start going down again. 813 00:46:01,900 --> 00:46:04,810 What happens to electricity growth? 814 00:46:04,810 --> 00:46:07,540 Is it above 4%? 815 00:46:07,540 --> 00:46:08,995 How many think it's above 4%? 816 00:46:11,600 --> 00:46:12,995 How many think it's below 4%? 817 00:46:15,560 --> 00:46:19,250 Some of you think it's an imaginary number, 818 00:46:19,250 --> 00:46:22,280 I'm missing votes here, OK. 819 00:46:22,280 --> 00:46:25,310 Those who voted say it's above 4%. 820 00:46:25,310 --> 00:46:26,450 And yet it wasn't. 821 00:46:30,170 --> 00:46:37,760 So now you say, how are people making these decisions? 822 00:46:37,760 --> 00:46:40,250 There's clearly some change that's 823 00:46:40,250 --> 00:46:42,400 happened on the demand side, right? 824 00:46:42,400 --> 00:46:47,210 I mean, something's different here. 825 00:46:47,210 --> 00:46:52,530 Is it that people were shocked by the high prices they 826 00:46:52,530 --> 00:46:55,260 encountered and the price increases they encountered 827 00:46:55,260 --> 00:46:59,220 in the '70s and for the first time they thought about not 828 00:46:59,220 --> 00:47:01,470 buying every electrical device? 829 00:47:01,470 --> 00:47:03,510 And so you got slower growth, even 830 00:47:03,510 --> 00:47:11,010 though incomes were growing and prices were declining a bit-- 831 00:47:11,010 --> 00:47:12,390 reasonable hypothesis. 832 00:47:15,500 --> 00:47:17,130 You can test it. 833 00:47:17,130 --> 00:47:23,670 Is there any price and income effect? 834 00:47:23,670 --> 00:47:26,840 Growth picks up in the '90s. 835 00:47:26,840 --> 00:47:30,740 Prices come down in the '90s for electricity and other things 836 00:47:30,740 --> 00:47:32,550 as well. 837 00:47:32,550 --> 00:47:34,330 What happens to electricity demand? 838 00:47:34,330 --> 00:47:35,580 Now, you've got to vote folks. 839 00:47:35,580 --> 00:47:38,940 This is no fun watching people go, oh, geez, I don't know. 840 00:47:38,940 --> 00:47:42,580 How many think its above 3%? 841 00:47:42,580 --> 00:47:46,480 OK, how many people think is below 3%? 842 00:47:46,480 --> 00:47:49,360 OK, the smart people win. 843 00:47:49,360 --> 00:47:51,580 It's below 3%. 844 00:47:51,580 --> 00:47:53,920 Why is it below 3%? 845 00:47:53,920 --> 00:47:55,480 Income growth picked up. 846 00:47:55,480 --> 00:47:58,940 Prices started declining pretty reasonably. 847 00:47:58,940 --> 00:47:59,954 No one quite knows. 848 00:47:59,954 --> 00:48:00,454 You know. 849 00:48:00,454 --> 00:48:02,070 AUDIENCE: A little earlier on in the years 850 00:48:02,070 --> 00:48:04,410 people were buying more appliances like refrigerators 851 00:48:04,410 --> 00:48:05,310 and TVs. 852 00:48:05,310 --> 00:48:07,170 And then as you get into the later years, 853 00:48:07,170 --> 00:48:08,970 everybody already has those appliances. 854 00:48:08,970 --> 00:48:10,667 But they were getting more efficient. 855 00:48:10,667 --> 00:48:12,750 And so they're replacing their existing appliances 856 00:48:12,750 --> 00:48:14,830 with more efficient appliances and so 857 00:48:14,830 --> 00:48:17,020 electricity use didn't grow. 858 00:48:17,020 --> 00:48:20,080 RICHARD SCHMALENSEE: Maybe, maybe, yeah, it's 859 00:48:20,080 --> 00:48:21,880 not implausible. 860 00:48:21,880 --> 00:48:23,320 You could test it, right? 861 00:48:23,320 --> 00:48:26,380 You could look at purchase patterns of various appliances. 862 00:48:26,380 --> 00:48:28,990 This is national use. 863 00:48:28,990 --> 00:48:31,720 You're getting the impact of probably a decline 864 00:48:31,720 --> 00:48:33,160 of manufacturing. 865 00:48:33,160 --> 00:48:34,420 This isn't all residential. 866 00:48:34,420 --> 00:48:36,730 So you're getting less industrial load. 867 00:48:36,730 --> 00:48:41,480 And you may also be getting more efficient new construction 868 00:48:41,480 --> 00:48:43,570 in office buildings. 869 00:48:43,570 --> 00:48:45,453 People began to think about efficiency. 870 00:48:45,453 --> 00:48:47,120 They didn't think much about efficiency. 871 00:48:47,120 --> 00:48:49,300 But when you start to think about efficiency, 872 00:48:49,300 --> 00:48:52,350 you begin to get gains in efficiency. 873 00:48:52,350 --> 00:48:54,430 So I'm guessing its efficiency. 874 00:48:54,430 --> 00:48:59,650 It's not that people stopped wanting energy services, 875 00:48:59,650 --> 00:49:03,160 but it's kind of implausible that suddenly you 876 00:49:03,160 --> 00:49:06,280 know, oh, we're going to go back to small black and white 877 00:49:06,280 --> 00:49:08,200 televisions because the big colors use 878 00:49:08,200 --> 00:49:09,550 too much electricity. 879 00:49:09,550 --> 00:49:12,160 And I'm willing to bet refrigerators 880 00:49:12,160 --> 00:49:18,840 didn't shrink over this period, so probably efficiency. 881 00:49:18,840 --> 00:49:22,080 But I can tell you, it's a lot easier 882 00:49:22,080 --> 00:49:24,870 to talk about it after the fact than it 883 00:49:24,870 --> 00:49:28,290 was to see it, to see it coming on the basis 884 00:49:28,290 --> 00:49:30,460 of historical experience. 885 00:49:30,460 --> 00:49:34,410 Oh, I should take this after '07. 886 00:49:34,410 --> 00:49:36,960 So this was not a great decade. 887 00:49:36,960 --> 00:49:40,065 And I won't trouble you more. 888 00:49:40,065 --> 00:49:41,190 This is not a great decade. 889 00:49:41,190 --> 00:49:42,270 Income growth slowed. 890 00:49:42,270 --> 00:49:46,540 And if I took it to the end of the decade, it's really ugly. 891 00:49:46,540 --> 00:49:48,210 Let me not run into the recession. 892 00:49:48,210 --> 00:49:50,940 But even before the recession, income growth slowed. 893 00:49:50,940 --> 00:49:52,960 Energy prices began to go up. 894 00:49:52,960 --> 00:49:57,540 And as you might expect, electricity use slowed as well. 895 00:49:57,540 --> 00:50:02,340 Forecasts out to about 20, 30 show less than 1% 896 00:50:02,340 --> 00:50:04,030 growth a year. 897 00:50:04,030 --> 00:50:07,960 So the business has changed from one where-- and even 898 00:50:07,960 --> 00:50:10,090 with reasonable economic growth assumptions-- 899 00:50:10,090 --> 00:50:16,420 where 10% and bigger than that before 1950, 5%, 10% 900 00:50:16,420 --> 00:50:18,640 to maybe 1%. 901 00:50:18,640 --> 00:50:22,095 Maybe 1% over the next 20 years, OK? 902 00:50:25,350 --> 00:50:27,720 That should have come up earlier. 903 00:50:27,720 --> 00:50:29,190 That's one caution. 904 00:50:29,190 --> 00:50:31,560 Demand functions are not constants of nature. 905 00:50:31,560 --> 00:50:36,780 And if anybody says the elasticity is x, 906 00:50:36,780 --> 00:50:39,420 it probably means the elasticity has been x. 907 00:50:39,420 --> 00:50:41,460 Or we've got a pretty good study that estimates 908 00:50:41,460 --> 00:50:42,870 that the elasticity is x. 909 00:50:42,870 --> 00:50:47,190 These are not constants of nature. 910 00:50:47,190 --> 00:50:53,070 And now let me sort of lead into the intro to Wednesday. 911 00:50:57,360 --> 00:51:05,760 The earlier discussion of derived demand and long run, 912 00:51:05,760 --> 00:51:08,820 short run and how you estimate sort of 913 00:51:08,820 --> 00:51:11,460 assumed in the background the standard economist model 914 00:51:11,460 --> 00:51:13,770 of rational behavior. 915 00:51:13,770 --> 00:51:14,850 Comes now, McKinsey. 916 00:51:17,440 --> 00:51:21,040 And you've got a very short McKinsey assignment 917 00:51:21,040 --> 00:51:22,690 on the readings. 918 00:51:22,690 --> 00:51:26,530 McKinsey has made a lot of money, basically country 919 00:51:26,530 --> 00:51:32,540 to country saying, how much energy conservation 920 00:51:32,540 --> 00:51:33,860 can be free? 921 00:51:33,860 --> 00:51:35,750 How can you do it for free? 922 00:51:35,750 --> 00:51:37,430 And they do curves like this. 923 00:51:37,430 --> 00:51:38,990 I forget what this one is for. 924 00:51:38,990 --> 00:51:42,870 But this is potential gigatons of carbon. 925 00:51:42,870 --> 00:51:44,390 But they do it for energy, as well. 926 00:51:44,390 --> 00:51:47,150 Gigatons of carbon dioxide reductions 927 00:51:47,150 --> 00:51:49,220 per year and the cost. 928 00:51:49,220 --> 00:51:51,350 And this is all negative. 929 00:51:51,350 --> 00:51:54,530 This is all better than free. 930 00:51:54,530 --> 00:51:56,460 And they have it for energy consumption. 931 00:51:56,460 --> 00:51:58,820 And they have it for a bunch of countries. 932 00:51:58,820 --> 00:52:02,000 As I say, they've made a lot of money. 933 00:52:02,000 --> 00:52:03,250 But they're not the only ones. 934 00:52:03,250 --> 00:52:07,990 Let me not pick on McKinsey, as much fun as that is to do. 935 00:52:07,990 --> 00:52:10,330 This is called the efficiency paradox. 936 00:52:10,330 --> 00:52:13,550 And it's been around since the '70s 937 00:52:13,550 --> 00:52:16,490 that there are lots of studies that say, look, 938 00:52:16,490 --> 00:52:19,220 you would save money if. 939 00:52:19,220 --> 00:52:23,975 Now, we're going to take a look at reasons why this persists. 940 00:52:27,460 --> 00:52:32,820 But I will give you an example I encountered personally. 941 00:52:32,820 --> 00:52:36,300 I came back from Washington, EPA had just started its Energy 942 00:52:36,300 --> 00:52:37,860 Star program. 943 00:52:37,860 --> 00:52:42,360 I went to people at MIT, and I said, why not sign up for it? 944 00:52:42,360 --> 00:52:45,330 Why doesn't MIT sign up for the Energy Star program because A, 945 00:52:45,330 --> 00:52:47,130 you get PR points. 946 00:52:47,130 --> 00:52:52,470 And B, all you promised to do is to use more efficient lighting 947 00:52:52,470 --> 00:52:55,210 when it saves you money. 948 00:52:55,210 --> 00:52:57,580 You promise to use more efficient lighting 949 00:52:57,580 --> 00:52:59,960 when it saves you money. 950 00:52:59,960 --> 00:53:03,940 I could not get that sale made. 951 00:53:03,940 --> 00:53:05,647 And nobody's behaving irrationally, 952 00:53:05,647 --> 00:53:06,730 and nobody's being a jerk. 953 00:53:06,730 --> 00:53:09,340 And we will come back to it later when 954 00:53:09,340 --> 00:53:12,760 we talk about biodiesel at MIT. 955 00:53:12,760 --> 00:53:15,760 Because it's exactly the same phenomenon. 956 00:53:15,760 --> 00:53:21,400 Why did it take them three years to sell biodiesel at MIT. 957 00:53:21,400 --> 00:53:28,740 So we'll talk about that set of reasons why this persists. 958 00:53:28,740 --> 00:53:31,560 But let me talk about some others. 959 00:53:31,560 --> 00:53:35,250 I mean, does anyone know who Amory Lovins is? 960 00:53:38,190 --> 00:53:41,610 OK, Amory Lovins has been an efficiency guru, 961 00:53:41,610 --> 00:53:46,750 an energy efficiency guru since the 1970s. 962 00:53:46,750 --> 00:53:50,500 Amory Lovins, he heads the Rocky Mountain Institute. 963 00:53:50,500 --> 00:53:54,550 He lives in a house that has quintuple pane glass and highly 964 00:53:54,550 --> 00:53:55,120 insulated. 965 00:53:55,120 --> 00:53:58,060 And I think uses no outside energy. 966 00:53:58,060 --> 00:53:59,470 And Amory, yeah, you know Amory. 967 00:53:59,470 --> 00:54:00,200 OK. 968 00:54:00,200 --> 00:54:02,530 AUDIENCE: Yeah, well, He has a super efficient home. 969 00:54:02,530 --> 00:54:05,270 But he can't run the computer off solar power 970 00:54:05,270 --> 00:54:08,020 so he does actually-- office equipment 971 00:54:08,020 --> 00:54:09,695 he has to get outside help for. 972 00:54:09,695 --> 00:54:12,070 RICHARD SCHMALENSEE: I'm really disappointed right there. 973 00:54:12,070 --> 00:54:15,160 Yeah, I'll tell you. 974 00:54:15,160 --> 00:54:20,030 But I was once lobbied by him in Washington. 975 00:54:20,030 --> 00:54:23,890 And he came in and explained how there 976 00:54:23,890 --> 00:54:27,940 are all these triple-pane glass and all these wonderful energy 977 00:54:27,940 --> 00:54:31,360 saving devices that would more than pay for themselves 978 00:54:31,360 --> 00:54:34,160 and reduce US energy consumption. 979 00:54:34,160 --> 00:54:38,810 And a guy who was working for me looked at him and said, 980 00:54:38,810 --> 00:54:42,530 you are saying consumers save money doing this stuff. 981 00:54:42,530 --> 00:54:44,930 He said, yes, that's right. 982 00:54:44,930 --> 00:54:47,173 And my guy's response was, that's fabulous. 983 00:54:47,173 --> 00:54:49,340 It means we don't have to have a government program. 984 00:54:52,400 --> 00:54:55,820 There was a pause. 985 00:54:55,820 --> 00:54:57,590 We'll come back to again. 986 00:54:57,590 --> 00:54:59,510 There ought to be a pause. 987 00:54:59,510 --> 00:55:01,610 But it was a great line. 988 00:55:01,610 --> 00:55:02,660 It was a great line. 989 00:55:02,660 --> 00:55:04,490 He hadn't thought of that. 990 00:55:04,490 --> 00:55:06,590 He has by now. 991 00:55:06,590 --> 00:55:08,570 So why? 992 00:55:08,570 --> 00:55:13,610 Well, one set of answers in the '70s was these. 993 00:55:13,610 --> 00:55:15,997 And I don't mean to demean engineers 994 00:55:15,997 --> 00:55:17,330 by saying engineering estimates. 995 00:55:17,330 --> 00:55:20,630 They're sort of back of the envelope, typical building, 996 00:55:20,630 --> 00:55:24,020 typical household kind of analysis 997 00:55:24,020 --> 00:55:27,860 that a lot of the early work was just too optimistic. 998 00:55:27,860 --> 00:55:30,327 It just assumed, you know, you'd throw out the furnace. 999 00:55:30,327 --> 00:55:31,910 Well, you don't throw out the furnace. 1000 00:55:31,910 --> 00:55:33,577 And you don't save money by throwing out 1001 00:55:33,577 --> 00:55:36,620 a five-year-old furnace and replacing it early. 1002 00:55:36,620 --> 00:55:44,440 So there were a lot of critiques that said, 1003 00:55:44,440 --> 00:55:46,540 these estimates are flawed. 1004 00:55:46,540 --> 00:55:48,280 Well, they might be flawed. 1005 00:55:48,280 --> 00:55:51,310 But there have been lots of clear examples later. 1006 00:55:51,310 --> 00:55:55,210 The MIT lighting example was an absolutely dead clear example 1007 00:55:55,210 --> 00:56:00,080 of money on the sidewalk, money on the sidewalk. 1008 00:56:00,080 --> 00:56:02,080 So the note sites, and I think I put 1009 00:56:02,080 --> 00:56:05,770 on the list the Hausman-Joskow paper. 1010 00:56:05,770 --> 00:56:10,940 And I say, well, if you look at consumer decision making-- 1011 00:56:10,940 --> 00:56:14,900 and Hausman did a classic study of air conditioners-- consumers 1012 00:56:14,900 --> 00:56:18,030 act like they have very high discount rates. 1013 00:56:18,030 --> 00:56:21,230 So you say, can you rationalize why consumers 1014 00:56:21,230 --> 00:56:24,110 don't spend a little more for energy efficient air 1015 00:56:24,110 --> 00:56:24,710 conditioners? 1016 00:56:24,710 --> 00:56:27,650 This was using data well before they were all really 1017 00:56:27,650 --> 00:56:29,153 efficient with standards. 1018 00:56:29,153 --> 00:56:30,320 How do you rationalize that? 1019 00:56:30,320 --> 00:56:32,660 Well, you rationalize it by saying, 1020 00:56:32,660 --> 00:56:36,230 consumers have limited access to capital. 1021 00:56:36,230 --> 00:56:39,400 They don't want to borrow to buy an air conditioner. 1022 00:56:39,400 --> 00:56:45,850 So they act as if their discount rate is 20%, 30% a year, 1023 00:56:45,850 --> 00:56:47,600 which means another way to think about it, 1024 00:56:47,600 --> 00:56:51,230 I need a really quick payback if I'm going to put money 1025 00:56:51,230 --> 00:56:54,200 up front in energy efficiency. 1026 00:56:54,200 --> 00:56:58,750 Well, two points, and I'll come to the second point 1027 00:56:58,750 --> 00:57:01,050 in just a minute. 1028 00:57:01,050 --> 00:57:08,650 The first point is, that doesn't explain business behavior. 1029 00:57:08,650 --> 00:57:15,160 The 20% to 30% discount rate, why would a business do that? 1030 00:57:15,160 --> 00:57:17,020 If you can raise capital at let's say 1031 00:57:17,020 --> 00:57:20,860 a canonical rate in the old days of 10%, 1032 00:57:20,860 --> 00:57:24,010 and you got an investment in efficiency that 1033 00:57:24,010 --> 00:57:28,830 has a 20% return, why wouldn't you make it? 1034 00:57:28,830 --> 00:57:31,330 You can see the household might have trouble raising capital 1035 00:57:31,330 --> 00:57:33,730 to insulate the house or to buy air conditioning, 1036 00:57:33,730 --> 00:57:35,470 to buy a more efficient air conditioner. 1037 00:57:35,470 --> 00:57:38,440 But Walmart, why wouldn't Walmart do it? 1038 00:57:38,440 --> 00:57:39,670 Well, Walmart probably does. 1039 00:57:39,670 --> 00:57:42,640 But why did we seem to see money lying 1040 00:57:42,640 --> 00:57:45,880 on the ground for commercial and industrial demand? 1041 00:57:45,880 --> 00:57:49,870 And we'll see a case later in the course of an energy saving 1042 00:57:49,870 --> 00:57:52,780 opportunity being confronted by a perfectly rational profit 1043 00:57:52,780 --> 00:57:55,360 maximizing firm, high rate of return that 1044 00:57:55,360 --> 00:57:58,042 is a very hard sell. 1045 00:57:58,042 --> 00:58:01,170 And we'll see biodiesel was a very hard sell. 1046 00:58:01,170 --> 00:58:03,270 So you got to have other explanations. 1047 00:58:03,270 --> 00:58:06,900 The high discount rate doesn't make sense for businesses. 1048 00:58:06,900 --> 00:58:07,733 Yeah. 1049 00:58:07,733 --> 00:58:09,730 AUDIENCE: Yes, would it make more sense 1050 00:58:09,730 --> 00:58:12,130 for the utility firms to kind of take 1051 00:58:12,130 --> 00:58:16,300 charge of going and trying to make buildings 1052 00:58:16,300 --> 00:58:18,868 more efficient in that sense? 1053 00:58:18,868 --> 00:58:21,160 They're going to be more aligned in terms of the profit 1054 00:58:21,160 --> 00:58:23,998 they can make because they'll be using less fuel 1055 00:58:23,998 --> 00:58:25,790 for the same sort of benefit they are going 1056 00:58:25,790 --> 00:58:26,873 to provide for the people. 1057 00:58:26,873 --> 00:58:29,980 RICHARD SCHMALENSEE: Well, we're going to come back to that. 1058 00:58:29,980 --> 00:58:34,180 It depends on, to get to that, you 1059 00:58:34,180 --> 00:58:37,010 have to have a diagnosis of the problem. 1060 00:58:37,010 --> 00:58:45,490 So if the problem is ignorance, which is a common argument, 1061 00:58:45,490 --> 00:58:48,700 then you say the utility has better information. 1062 00:58:48,700 --> 00:58:50,560 And so it could probably do this. 1063 00:58:50,560 --> 00:58:52,720 Although, the utilities expertise 1064 00:58:52,720 --> 00:58:56,350 is generating electricity, not cooling buildings, right? 1065 00:58:56,350 --> 00:59:01,520 They can acquire the expertise, but so could some third party. 1066 00:59:01,520 --> 00:59:02,630 And some firms do, right? 1067 00:59:02,630 --> 00:59:04,940 People do go around. 1068 00:59:04,940 --> 00:59:07,130 There are a number of firms, I'm involved 1069 00:59:07,130 --> 00:59:10,490 with one of them that is trying to make a business out 1070 00:59:10,490 --> 00:59:13,900 of selling energy efficiency to households. 1071 00:59:13,900 --> 00:59:17,530 It is not an easy business, not an easy business. 1072 00:59:17,530 --> 00:59:20,110 But the problem with utilities though-- 1073 00:59:20,110 --> 00:59:22,300 and we'll come back to this for sure-- 1074 00:59:22,300 --> 00:59:25,670 they get compensated for selling electricity. 1075 00:59:25,670 --> 00:59:31,090 So they have an inherent conflict. 1076 00:59:31,090 --> 00:59:35,240 If I can find a way not to rely on somebody that's conflicted, 1077 00:59:35,240 --> 00:59:36,490 I will. 1078 00:59:36,490 --> 00:59:40,692 If you say, I'm paying you for selling electricity now, 1079 00:59:40,692 --> 00:59:42,400 I'd like you to go around and tell people 1080 00:59:42,400 --> 00:59:46,750 how to consume less, I'd begin to wonder how effective you're 1081 00:59:46,750 --> 00:59:48,100 going to be. 1082 00:59:48,100 --> 00:59:49,000 I can make you do it. 1083 00:59:49,000 --> 00:59:50,042 I can make you go around. 1084 00:59:50,042 --> 00:59:52,510 And states do this, but the effectiveness 1085 00:59:52,510 --> 00:59:58,210 varies because your motivation isn't necessarily terrific. 1086 00:59:58,210 --> 01:00:02,140 So we've got this low hanging fruit problem. 1087 01:00:02,140 --> 01:00:06,040 And we're going to see this in a number of settings 1088 01:00:06,040 --> 01:00:10,510 that how come opportunities to increase efficiency 1089 01:00:10,510 --> 01:00:16,230 that are apparently profitable don't get picked up. 1090 01:00:16,230 --> 01:00:19,440 But the world is even more complicated. 1091 01:00:19,440 --> 01:00:22,100 Sometimes people over-invest. 1092 01:00:22,100 --> 01:00:23,840 Let me talk about the Prius. 1093 01:00:23,840 --> 01:00:26,750 How many people or their parents have 1094 01:00:26,750 --> 01:00:29,270 Prius, or relatives have Prius? 1095 01:00:29,270 --> 01:00:31,250 Yeah, lots of people have Prius. 1096 01:00:31,250 --> 01:00:33,470 And I'm not going to criticize your rationality 1097 01:00:33,470 --> 01:00:35,630 in what I'm saying here, OK? 1098 01:00:35,630 --> 01:00:36,560 OK. 1099 01:00:36,560 --> 01:00:38,780 So three million people have bought them 1100 01:00:38,780 --> 01:00:40,880 and they sold three million worldwide. 1101 01:00:40,880 --> 01:00:43,478 They've sold over a million in the US. 1102 01:00:43,478 --> 01:00:45,770 I'm now relying on others who have looked hard at this. 1103 01:00:45,770 --> 01:00:47,660 But they tell me that in terms of driving, 1104 01:00:47,660 --> 01:00:51,150 in terms of interior amenities, the styling is different. 1105 01:00:51,150 --> 01:00:55,710 But in terms of interior design, it's a Corolla. 1106 01:00:55,710 --> 01:00:58,260 But it's a Corolla that gets 50 miles a gallon 1107 01:00:58,260 --> 01:01:01,440 instead of 28 miles a gallon. 1108 01:01:01,440 --> 01:01:05,990 So let's ask the question, is it worth the difference 1109 01:01:05,990 --> 01:01:07,856 in upfront cost? 1110 01:01:07,856 --> 01:01:12,650 Did these three million people make a good Investment? 1111 01:01:12,650 --> 01:01:16,100 Remember, these are people who are not buying energy 1112 01:01:16,100 --> 01:01:17,930 efficient air conditioners. 1113 01:01:17,930 --> 01:01:21,420 Well, maybe they are, but not that many others are. 1114 01:01:21,420 --> 01:01:23,360 So let's look at it. 1115 01:01:23,360 --> 01:01:26,580 We're going to ask if the cost difference-- 1116 01:01:26,580 --> 01:01:28,080 so you save money on gasoline. 1117 01:01:28,080 --> 01:01:31,010 A cost difference is the upfront cost difference. 1118 01:01:31,010 --> 01:01:33,530 S is the annual savings on gasoline. 1119 01:01:33,530 --> 01:01:34,640 R is the interest rate. 1120 01:01:34,640 --> 01:01:37,400 And T is, how long do you have to keep 1121 01:01:37,400 --> 01:01:41,030 the car until the savings on gasoline 1122 01:01:41,030 --> 01:01:43,400 justify the purchase price. 1123 01:01:43,400 --> 01:01:47,000 OK, so that's just a discounting exercise. 1124 01:01:47,000 --> 01:01:53,420 If there's no interest rate, you just divide the upfront cost 1125 01:01:53,420 --> 01:01:57,360 by the annual savings and that gives you the number of years. 1126 01:01:57,360 --> 01:01:59,010 That's 0 interest rate. 1127 01:01:59,010 --> 01:02:01,620 If there is a positive interest rate, 1128 01:02:01,620 --> 01:02:06,120 then you ask, you discount S per year out to T. 1129 01:02:06,120 --> 01:02:07,980 And you say, how large does T have 1130 01:02:07,980 --> 01:02:13,790 to be so that discounted savings equals the original cost. 1131 01:02:13,790 --> 01:02:19,340 Clearly if the savings are small, 1132 01:02:19,340 --> 01:02:20,510 there can be no solution. 1133 01:02:20,510 --> 01:02:22,670 There can be no positive solution to this. 1134 01:02:22,670 --> 01:02:26,840 But if RC is less than S, it's this simple formula 1135 01:02:26,840 --> 01:02:30,710 you can get in less than 5 seconds. 1136 01:02:30,710 --> 01:02:32,450 So let's look at it. 1137 01:02:32,450 --> 01:02:36,680 Toyota, these are data a few years old. 1138 01:02:36,680 --> 01:02:40,580 And they don't count the subsidies 1139 01:02:40,580 --> 01:02:42,380 that were there for the first-- 1140 01:02:42,380 --> 01:02:45,230 in the US I think it was 60,000 units sold. 1141 01:02:45,230 --> 01:02:47,940 But the subsidies have long since run out. 1142 01:02:47,940 --> 01:02:52,240 So the average premium, 2008, maybe was $7,450. 1143 01:02:52,240 --> 01:02:53,790 Assume maintenance is the same. 1144 01:02:53,790 --> 01:02:54,950 It may not be. 1145 01:02:54,950 --> 01:02:57,740 Prius may be cheaper. 1146 01:02:57,740 --> 01:03:00,800 Car's average about 12,000 miles a year. 1147 01:03:00,800 --> 01:03:03,050 I can work out the annual cost. 1148 01:03:03,050 --> 01:03:08,530 The difference, the savings is $566. 1149 01:03:08,530 --> 01:03:16,350 So if the interest rate is 0, you've 1150 01:03:16,350 --> 01:03:20,110 got to keep the car for 13.2 years to break even. 1151 01:03:20,110 --> 01:03:23,280 Now again, if you have a subsidy or if the gasoline 1152 01:03:23,280 --> 01:03:27,180 is more expensive, this will get smaller. 1153 01:03:27,180 --> 01:03:29,430 But that's kind of a long time. 1154 01:03:29,430 --> 01:03:33,570 Cars tend to turn over after about 10, really. 1155 01:03:33,570 --> 01:03:36,210 And most people don't say, hold it for 10. 1156 01:03:36,210 --> 01:03:38,610 10 is when they go out of the fleet. 1157 01:03:38,610 --> 01:03:40,950 So if you're going to resell, you've 1158 01:03:40,950 --> 01:03:45,180 got to sell to somebody who also loves Prius. 1159 01:03:45,180 --> 01:03:50,550 If the interest rate is 5%, not that high, 1160 01:03:50,550 --> 01:03:53,850 you've got to hold the car for 21.6 years 1161 01:03:53,850 --> 01:03:57,880 before those gasoline savings pay for the cost difference. 1162 01:03:57,880 --> 01:04:02,680 If the interest rate is above 7.6%, there is no break even. 1163 01:04:02,680 --> 01:04:07,640 You can verify that you don't satisfy that. 1164 01:04:07,640 --> 01:04:13,090 So you can hold it forever and the discounted gasoline savings 1165 01:04:13,090 --> 01:04:18,547 don't equal the original purchase. 1166 01:04:18,547 --> 01:04:19,630 Now, you can play with it. 1167 01:04:19,630 --> 01:04:21,790 You can put in a per vehicle subsidy. 1168 01:04:21,790 --> 01:04:24,910 You can make gasoline more expensive. 1169 01:04:24,910 --> 01:04:28,420 But these don't seem like crazy numbers 1170 01:04:28,420 --> 01:04:30,190 in light of how those cars work. 1171 01:04:37,040 --> 01:04:40,590 And this is the puzzle. 1172 01:04:40,590 --> 01:04:42,960 People act when they're buying air conditioners 1173 01:04:42,960 --> 01:04:46,150 like their discount rate is 20% to 30%. 1174 01:04:46,150 --> 01:04:49,650 And they act when they're buying a Prius like they have 1175 01:04:49,650 --> 01:04:53,790 a really low discount rate. 1176 01:04:53,790 --> 01:04:58,040 Or maybe this is not the right way to look at it. 1177 01:05:00,870 --> 01:05:03,030 Tell no other economists I said this. 1178 01:05:03,030 --> 01:05:06,960 But it may be that this sort of economic approach I was 1179 01:05:06,960 --> 01:05:08,970 taking here is missing stuff. 1180 01:05:12,230 --> 01:05:13,190 It's missing a lot. 1181 01:05:13,190 --> 01:05:15,430 Why do people buy a Prius by the way? 1182 01:05:15,430 --> 01:05:18,000 Do people buy a Prius to save gasoline? 1183 01:05:18,000 --> 01:05:18,500 Scot. 1184 01:05:18,500 --> 01:05:20,917 AUDIENCE: A lot of people just buy environmental friendly. 1185 01:05:20,917 --> 01:05:23,430 And they want to do something and something like that. 1186 01:05:23,430 --> 01:05:26,340 And so even if it might not be saving them that much money 1187 01:05:26,340 --> 01:05:28,560 or saving them money at all, they still do it. 1188 01:05:28,560 --> 01:05:32,620 RICHARD SCHMALENSEE: They want to do the right thing. 1189 01:05:32,620 --> 01:05:34,140 There's another argument that people 1190 01:05:34,140 --> 01:05:36,780 make is, there's a reason the Prius doesn't look 1191 01:05:36,780 --> 01:05:41,460 like the Corolla, so not only do you feel satisfied that you're 1192 01:05:41,460 --> 01:05:45,120 doing a good thing, people can see you're doing a good thing. 1193 01:05:45,120 --> 01:05:48,900 Because I mean, that's a funny looking car to be honest. 1194 01:05:48,900 --> 01:05:51,270 And people know you're driving a Prius. 1195 01:05:51,270 --> 01:05:53,550 There are other hybrids you don't people are driving. 1196 01:05:53,550 --> 01:05:56,890 And they don't sell as well as the Prius. 1197 01:05:56,890 --> 01:05:58,270 None of that's here. 1198 01:05:58,270 --> 01:06:00,850 The do the good thing isn't there. 1199 01:06:00,850 --> 01:06:04,960 The other people see I'm doing a good thing isn't there. 1200 01:06:04,960 --> 01:06:10,840 This is not the drive demand for transportation services story 1201 01:06:10,840 --> 01:06:13,260 that gets the Prius sold. 1202 01:06:13,260 --> 01:06:14,295 This is other things. 1203 01:06:17,080 --> 01:06:22,330 And we're going to hear from professor Silbey on Wednesday 1204 01:06:22,330 --> 01:06:24,040 about some of the other things. 1205 01:06:24,040 --> 01:06:26,470 Have you questions or comments. 1206 01:06:26,470 --> 01:06:28,150 Paper going OK. 1207 01:06:28,150 --> 01:06:29,352 Obena. 1208 01:06:29,352 --> 01:06:31,810 AUDIENCE: Did they consider the time saved at the gas pump? 1209 01:06:31,810 --> 01:06:34,220 RICHARD SCHMALENSEE: Hmm, time saved at the gas pump? 1210 01:06:34,220 --> 01:06:35,800 AUDIENCE: Right. 1211 01:06:35,800 --> 01:06:37,550 RICHARD SCHMALENSEE: I don't know the size 1212 01:06:37,550 --> 01:06:38,930 of the tank on that thing. 1213 01:06:38,930 --> 01:06:40,070 Does it save you time? 1214 01:06:40,070 --> 01:06:41,690 It probably does. 1215 01:06:41,690 --> 01:06:44,420 AUDIENCE: Full time, so that would be two or three weeks. 1216 01:06:44,420 --> 01:06:45,020 RICHARD SCHMALENSEE: These are people who 1217 01:06:45,020 --> 01:06:46,490 value their time very highly. 1218 01:06:49,430 --> 01:06:51,683 I have to say, I value mine really highly. 1219 01:06:51,683 --> 01:06:52,850 And it never occurred to me. 1220 01:06:52,850 --> 01:06:53,900 But it could be. 1221 01:06:53,900 --> 01:06:56,420 It could be you people all bill out at $1,000 an hour. 1222 01:06:56,420 --> 01:06:58,580 And a few minutes is worth saving. 1223 01:06:58,580 --> 01:07:00,140 Any other questions, comments. 1224 01:07:00,140 --> 01:07:01,880 David. 1225 01:07:01,880 --> 01:07:03,220 AUDIENCE: Yes. 1226 01:07:03,220 --> 01:07:04,850 You could also regulate a small town 1227 01:07:04,850 --> 01:07:07,190 than the average driven mile? 1228 01:07:07,190 --> 01:07:10,535 So somebody drives less in 2003. 1229 01:07:10,535 --> 01:07:12,160 RICHARD SCHMALENSEE: That might be low. 1230 01:07:12,160 --> 01:07:13,510 So maybe it's segments that way. 1231 01:07:13,510 --> 01:07:14,920 Also testable. 1232 01:07:14,920 --> 01:07:17,100 Sarah, does your family drive a lot? 1233 01:07:17,100 --> 01:07:19,826 AUDIENCE: No, what about risk aversion and price fluctuations 1234 01:07:19,826 --> 01:07:21,360 in gas? 1235 01:07:21,360 --> 01:07:23,860 RICHARD SCHMALENSEE: You figure this is kind of an insurance 1236 01:07:23,860 --> 01:07:26,590 policy that when gas prices move, we won't be hit too hard. 1237 01:07:26,590 --> 01:07:29,967 That's another possibility, also testable, yeah. 1238 01:07:29,967 --> 01:07:32,300 AUDIENCE: I read an article that production of the Prius 1239 01:07:32,300 --> 01:07:34,396 is actually more inefficient than the production 1240 01:07:34,396 --> 01:07:35,520 of the Corolla. 1241 01:07:35,520 --> 01:07:38,816 And the energy that people claimed 1242 01:07:38,816 --> 01:07:42,852 they are saving actually because of the inefficiency 1243 01:07:42,852 --> 01:07:47,130 of production, you know, they established it doesn't actually 1244 01:07:47,130 --> 01:07:48,020 save energy. 1245 01:07:48,020 --> 01:07:48,890 RICHARD SCHMALENSEE: So they think 1246 01:07:48,890 --> 01:07:50,705 they're doing a good thing but they're not. 1247 01:07:53,490 --> 01:07:54,540 It's a terrible tragedy. 1248 01:07:54,540 --> 01:07:57,320 Bring your name cards next time.