WEBVTT NOTE The Praxeology of the Knowledge Problem of Socialism 1 00:00:00.000 --> 00:00:15.000 So I'm going to be discussing the praxeology of the knowledge problem of socialism and the reason I wrote this is that I feel like lots of Rothbardians reject the knowledge problem completely 2 00:00:15.000 --> 00:00:22.000 and I want to put the knowledge problem within the praxeological Rothbardian framework 3 00:00:22.000 --> 00:00:29.000 and I heard somebody ask earlier today what is praxeology a few minutes ago 4 00:00:29.000 --> 00:00:35.160 So, praxeology is the study of human action. Just read any, read human action and you're 5 00:00:35.160 --> 00:00:45.280 studying praxeology. So, let me say something about the two problems because if somebody 6 00:00:45.280 --> 00:00:49.600 doesn't know what praxeology is, I should probably explain the calculation problem and 7 00:00:49.600 --> 00:00:58.680 the knowledge problem too. So, the calculation problem refers to the fact that under socialism, 8 00:00:58.680 --> 00:01:19.680 It is not possible to give any objective sense of profit and loss for any given production process because all costs and benefits are given in terms of units that cannot be compared objectively, cannot be added up like apples and oranges. 9 00:01:19.680 --> 00:01:31.520 If I am running a labor camp, some of my costs will be in terms of the labor of my slaves. 10 00:01:31.520 --> 00:01:38.840 Other costs will be given in terms of money for things that I bought for materials production. 11 00:01:38.840 --> 00:01:45.520 So my costs will be money costs plus labor costs, but since I never bought the labor, 12 00:01:45.520 --> 00:01:50.820 The labor is not given in terms of any price, so there isn't any way to say objectively 13 00:01:50.820 --> 00:01:58.300 whether the labor and money costs add up to something that is worth the benefits. 14 00:01:58.300 --> 00:02:04.960 So probably the slaves would say that it's not worth the benefits, whereas the slave 15 00:02:04.960 --> 00:02:08.360 owner would say, yes, it is worth the benefits. 16 00:02:08.360 --> 00:02:10.900 So that's the calculation problem. 17 00:02:10.900 --> 00:02:20.500 Because the knowledge problem refers to the fact that if you have a large organization, 18 00:02:20.500 --> 00:02:30.340 you have a multinational corporation or a large state, you have all sorts of subsidiary 19 00:02:30.340 --> 00:02:35.740 organizations everywhere that are supposed to be doing what you want, you have to gain 20 00:02:35.740 --> 00:02:45.740 and a knowledge of everything that's happening to all of your subsidiary organizations to tell them what to do, otherwise how are you going to tell? 21 00:02:45.740 --> 00:02:52.740 So I'm going to start with a quote from Rothbard, Man Economy and State. 22 00:02:52.740 --> 00:03:02.740 He's referring to a free market organization that has completely, vertically integrated 23 00:03:02.740 --> 00:03:06.660 the production and use of a given capital good. 24 00:03:06.660 --> 00:03:12.020 So he says, suppose there is no external market, i.e. that the Jones Company is the only producer 25 00:03:12.020 --> 00:03:13.780 of the intermediate good. 26 00:03:13.780 --> 00:03:17.740 It would have no way of knowing which stage was being conducted profitably and which was 27 00:03:17.740 --> 00:03:18.740 not. 28 00:03:18.740 --> 00:03:23.740 In that case, it would have no way of knowing which stage was being conducted properly and which was not. 29 00:03:23.740 --> 00:03:28.740 It would therefore have no way of knowing how to allocate factors of the various stages. 30 00:03:28.740 --> 00:03:34.740 There would be no way for it to estimate any implicit price or opportunity cost for the capital good at that particular stage. 31 00:03:34.740 --> 00:03:40.740 Any estimate would be completely arbitrary and have no meaningful relation to economic conditions. 32 00:03:40.740 --> 00:03:49.740 Not being able to calculate a price, a firm could not rationally allocate factors and resources from one stage to another. 33 00:03:49.740 --> 00:03:54.740 So I believe that Rothbard has made a mistake here. 34 00:03:54.740 --> 00:04:07.740 Now one of the arguments that is often used against the Hayekian position is that Hayek will say that prices communicate knowledge and they tell us, tell entrepreneurs what they should do. 35 00:04:07.740 --> 00:04:13.740 Well, that's actually incorrect because if you want to know what's going to be profitable tomorrow, 36 00:04:13.740 --> 00:04:17.740 the prices of today tell you nothing about that. 37 00:04:17.740 --> 00:04:20.740 You really need to know about what people's preference scales are. 38 00:04:20.740 --> 00:04:25.740 You need to actually learn about what people really want if you're going to produce things. 39 00:04:25.740 --> 00:04:30.740 So it's really preference scales that matter in terms of appraising a good. 40 00:04:30.740 --> 00:04:46.740 and on the free market it should be possible to appraise a good even if it's not being traded today, it might be traded tomorrow, if you know about people's preference scales that should be feasible in theory. 41 00:04:46.740 --> 00:04:55.740 So I think that the argument that is used against Hayek could really be used against Rothbard here as well. 42 00:04:55.740 --> 00:05:05.020 So, now here's sort of a diagram of a two-stage production process, and in the middle would 43 00:05:05.020 --> 00:05:07.820 be the vertically integrated capital good. 44 00:05:07.820 --> 00:05:13.060 I'm going to explain how we can appraise the vertically integrated capital good now. 45 00:05:13.060 --> 00:05:22.180 Basically, so I might have a choice of whether I want to invest more and get a second of 46 00:05:22.180 --> 00:05:27.920 of these capital goods, then I can produce more output, or whether it's more profitable 47 00:05:27.920 --> 00:05:31.800 just to produce the lower amount of output. 48 00:05:31.800 --> 00:05:39.460 So what I can do if I'm the manager of this company, the inputs are all on a market, so 49 00:05:39.460 --> 00:05:45.100 they have prices and I can appraise those just as I would with anything else. 50 00:05:45.100 --> 00:05:50.100 The outputs also have prices and I can appraise them like I would with anything else. 51 00:05:50.100 --> 00:05:56.620 So any kind of technological possibility for the intermediate vertically integrated capital 52 00:05:56.620 --> 00:06:05.820 good, I can just look at the overall profitability for the entire production process and figure 53 00:06:05.820 --> 00:06:09.500 out which use of this capital good is best. 54 00:06:09.500 --> 00:06:18.740 So do I want to invest more and rely more heavily on capital or do I want to stay how 55 00:06:18.740 --> 00:06:19.740 I am? 56 00:06:19.740 --> 00:06:25.940 And I can just look at the overall profitability of the entire production process. 57 00:06:25.940 --> 00:06:31.700 And this is really no different than what any entrepreneur does with a single production 58 00:06:31.700 --> 00:06:32.700 process. 59 00:06:32.700 --> 00:06:40.300 You're just looking at the optimal technological possibilities for a given, for the given process. 60 00:06:40.300 --> 00:06:47.900 It's just that in this case, there is a larger range of possibilities, there's just more 61 00:06:47.900 --> 00:06:53.260 or to consider. 62 00:06:53.260 --> 00:06:57.100 And you can even come up with a price for the intermediate capital good, even though 63 00:06:57.100 --> 00:06:58.940 there's no market. 64 00:06:58.940 --> 00:07:05.900 When you have the optimal production process, it must be that the intermediate good is in 65 00:07:05.900 --> 00:07:12.940 equilibrium with respect to the inputs and the outputs, necessarily. 66 00:07:12.940 --> 00:07:19.900 So you can assign it the price which equalizes the profitability of both stages of production. 67 00:07:19.900 --> 00:07:23.420 Now that is not a market price, so what does it mean? 68 00:07:23.420 --> 00:07:30.940 Well it has a very clear meaning on the market, and that is that I assign the intermediate 69 00:07:30.940 --> 00:07:37.660 capital good a price, now somebody else comes up to my business and says, you know, I will 70 00:07:37.660 --> 00:07:44.660 If I will buy this for a higher price than what I just calculated for, then I know that 71 00:07:44.660 --> 00:07:51.940 I should stop doing the second production process entirely and I should just focus on 72 00:07:51.940 --> 00:07:56.260 the first one and sell to the second guy, because it's more profitable just to focus 73 00:07:56.260 --> 00:07:57.700 on the first one. 74 00:07:57.700 --> 00:08:02.340 Or if somebody else starts producing this capital good for a lower price than what I've 75 00:08:02.340 --> 00:08:08.100 I've appraised it, then I know that I should specialize in the second production process 76 00:08:08.100 --> 00:08:11.660 and just focus on that and buy from him. 77 00:08:11.660 --> 00:08:17.780 So this appraised price has a real meaning even though it's not being traded on the market. 78 00:08:17.780 --> 00:08:24.940 Now let's consider, well first of all I think, notice that my argument does not rely on equilibrium 79 00:08:24.940 --> 00:08:31.100 assumptions, a limited number of inputs or stages of production, interpersonal comparisons 80 00:08:31.100 --> 00:08:38.460 is a value. It doesn't rely on anything that socialists like to sneak into their arguments 81 00:08:38.460 --> 00:08:45.300 to say that calculation under socialism is possible. There could be an infinite number 82 00:08:45.300 --> 00:08:50.160 of possible stages. There could be a lot more choices than the ones I've considered. I just 83 00:08:50.160 --> 00:08:59.900 listed two here, but there could be any number. But really, any entrepreneur has to contend 84 00:08:59.900 --> 00:09:04.580 with, hypothetically, an infinite number of possible technological choices. 85 00:09:04.580 --> 00:09:07.700 So there really isn't a difference here. 86 00:09:07.700 --> 00:09:17.420 Now, on the other hand, what if this intermediate capital good were not a free market good? 87 00:09:17.420 --> 00:09:25.300 Were not the case that I monopolize it just because I'm the best on the market, but because 88 00:09:25.300 --> 00:09:29.900 Because there's a monopoly and I'm granted a right from the state to say, you're the 89 00:09:29.900 --> 00:09:34.940 only one, nobody else is allowed to produce and use this good. 90 00:09:34.940 --> 00:09:40.580 Well then, although I can appraise the price for it that I mentioned earlier, then that 91 00:09:40.580 --> 00:09:46.660 price no longer has any economic meaning whatsoever because it could very well be that there are 92 00:09:46.660 --> 00:09:52.540 other people around who are perfectly willing to produce the good at a lower price or to 93 00:09:52.540 --> 00:09:57.260 by the good at a higher price and they can't do anything about it because 94 00:09:57.260 --> 00:10:02.860 they're prevented coercively so even though I can use this method to appraise 95 00:10:02.860 --> 00:10:11.740 a price on the free market and and allocate the good rationally in a 96 00:10:11.740 --> 00:10:18.980 coercive monopoly system the good would still be allocated incorrectly at least 97 00:10:18.980 --> 00:10:28.580 So I would say that what Rothbard was discussing in this part of Man Economy and State was 98 00:10:28.580 --> 00:10:32.780 not a calculation problem but a knowledge problem. 99 00:10:32.780 --> 00:10:41.220 It really depends on being able to think about two stages of production at the same time 100 00:10:41.220 --> 00:10:43.820 rather than just one. 101 00:10:43.820 --> 00:10:48.560 The issue is that if you had one manager managing the first stage of production and a different 102 00:10:48.560 --> 00:10:52.360 Management Manager managing the second stages of production, then they wouldn't know what 103 00:10:52.360 --> 00:10:53.360 to do. 104 00:10:53.360 --> 00:10:58.320 You would have to have one manager managing both stages of production, then he would be 105 00:10:58.320 --> 00:11:07.760 capable of appraising the good, the intermediate capital good, and arranging it correctly. 106 00:11:07.760 --> 00:11:17.800 So I will argue in my paper that there is not 107 00:11:17.800 --> 00:11:33.800 a calculation problem on the free market, Contra Rothbard. If there is potential for trade, there is a method to appraise goods and to allocate them rationally. There is only a calculation problem under socialism. 108 00:11:33.800 --> 00:11:45.560 So, okay, I was going to talk about... I talked about this part first. So here are some quotes 109 00:11:45.560 --> 00:11:54.720 from Mises from Bureaucracy, which is the book he wrote with the highest greatness density, 110 00:11:54.720 --> 00:12:02.740 in my opinion. And this, I think, the book Bureaucracy is really more about the knowledge 111 00:12:02.740 --> 00:12:10.520 problem than the calculation problem in a way because it really discusses a lot about 112 00:12:10.520 --> 00:12:19.220 how a socialist organization is going to get all of its subsidiary organizations to do 113 00:12:19.220 --> 00:12:20.360 what it wants. 114 00:12:20.360 --> 00:12:26.420 And this kind of explains the relationship between the calculation and knowledge problems 115 00:12:26.420 --> 00:12:28.760 under socialism. 116 00:12:28.760 --> 00:12:37.960 So under socialism there is a calculation problem because there is a coercion and this 117 00:12:37.960 --> 00:12:41.800 prevents goods from being appraised on the market. 118 00:12:41.800 --> 00:12:51.200 But what was I saying here? 119 00:12:51.200 --> 00:13:14.680 So what Mises says is that a free market firm is able to delegate authority because the 120 00:13:14.680 --> 00:13:21.040 subsidiary managers are able to measure profits and losses for their own individual parts 121 00:13:21.040 --> 00:13:24.960 Parts of the Production Process. 122 00:13:24.960 --> 00:13:31.520 Whereas a socialist organization, when they try to delegate authority, there is no such 123 00:13:31.520 --> 00:13:38.560 method that the subsidiary organizations are able to measure profit and loss. 124 00:13:38.560 --> 00:13:46.000 So whereas a free market organization can delegate authority and expect that the subsidiary 125 00:13:46.000 --> 00:14:14.220 Mises did not seem to take this argument to its logical extreme, but here I have a graph 126 00:14:14.220 --> 00:14:19.860 representing what is possible on the free market, but what would be necessary under 127 00:14:19.860 --> 00:14:25.020 socialism in order to find an optimum. 128 00:14:25.020 --> 00:14:32.060 So if you're a socialist and you're running an economy or attempting to, you can't expect 129 00:14:32.060 --> 00:14:35.500 any subsidiary organizations will do what you want. 130 00:14:35.500 --> 00:14:41.060 I mean, they might have sort of a guess, but there's no sense of any optimality that will 131 00:14:41.060 --> 00:14:49.060 will allow them to, that will give you any assurance that they're necessarily going to do what you want. 132 00:14:49.060 --> 00:14:58.060 So if you have perfect knowledge and you're able to look at the entire economy and just make all the decisions for yourself, 133 00:14:58.060 --> 00:15:05.060 that's what you will do. Whereas on the free market it is possible to have subsidiary organizations 134 00:15:05.060 --> 00:15:13.300 that will optimize their own profits and do what you want. Now obviously in real 135 00:15:13.300 --> 00:15:18.940 life this is not really feasible because people don't have perfect knowledge but 136 00:15:18.940 --> 00:15:25.620 what this shows is that there is a separate knowledge problem under 137 00:15:25.620 --> 00:15:31.700 socialism and it's caused by the calculation problem because there 138 00:15:31.700 --> 00:15:38.700 If there wasn't a calculation problem, you wouldn't really want to organize your economy in this way. 139 00:15:38.700 --> 00:15:45.700 But you can imagine an economy in which the knowledge problem has been solved, 140 00:15:45.700 --> 00:15:48.700 a socialist economy in which the knowledge problem has been solved, 141 00:15:48.700 --> 00:15:52.700 and compare it to one in which it has not been solved. 142 00:15:52.700 --> 00:15:56.700 And in my paper, I do this thought experiment, 143 00:15:56.700 --> 00:16:01.060 but I didn't really think I would have the time to go into all the details. 144 00:16:01.060 --> 00:16:05.300 So anyway, here's a summary. 145 00:16:05.300 --> 00:16:13.580 A bureaucratic organization can have no reason to expect that independent decisions made by subordinates will be beneficial to his overall plan. 146 00:16:13.580 --> 00:16:18.300 A free market organization need not centralize all knowledge and decisions, 147 00:16:18.300 --> 00:16:26.540 whereas a bureaucratic organization necessarily must if it wants to achieve any sort of optimum 148 00:16:26.540 --> 00:16:38.540 for its goals. The knowledge problem is caused by the calculation problem, but it's possible to compare socialist society, which has solved the knowledge problem, to one which has not. 149 00:16:38.540 --> 00:16:51.540 And here's the basic thesis of my paper. The calculation problem is a problem of evaluating a goal, the knowledge problem is the problem of actually achieving it. 150 00:16:51.540 --> 00:17:12.540 Now as I showed earlier, when you have a whole network of intermediate capital goods, if you can imagine the entire problem all at once, it is possible to allocate the capital goods according to your goal, but there won't be any overall idea of profit and loss. 151 00:17:12.540 --> 00:17:25.540 So if you do solve the knowledge problem, the ultimate problem of socialism is that the final goods that are produced will have no relation to people's preferences. 152 00:17:25.540 --> 00:17:37.540 But you wouldn't see problems like tractors rotting in the fields, huge piles of crops being produced that then just rot. 153 00:17:37.540 --> 00:17:52.540 So the Rothbardian interpretation, so Mises discussed the calculation and knowledge problems in his writings, but did not clearly distinguish them in points. 154 00:17:52.540 --> 00:18:01.540 The Rothbardian interpretation of Mises is characterized by failing to appreciate this distinction, 155 00:18:01.540 --> 00:18:12.540 and the Hayekian interpretation is characterized by not understanding the calculation problem at all because they only talk about knowledge. 156 00:18:12.540 --> 00:18:17.540 So here are some quotes from Mises to substantiate my position. 157 00:18:17.540 --> 00:18:29.540 The first step that would be necessary to form sections inside the socialist community to which the management of definite branches of business would be entrusted. 158 00:18:29.540 --> 00:18:35.540 The first step, which would be necessary, would be to form sections inside the socialist community 159 00:18:35.540 --> 00:18:38.540 to which the management of definite branches of business would be entrusted. 160 00:18:38.540 --> 00:18:42.540 As long as the industry of the socialist community is directed by one single authority, 161 00:18:42.540 --> 00:18:45.540 which makes all arrangements and bears all the responsibility, 162 00:18:45.540 --> 00:18:53.540 a solution of the problems is inconceivable because all the other workers are only acting instruments 163 00:18:53.540 --> 00:19:00.540 without independent delimited spheres of operation and consequently without any special responsibility. 164 00:19:00.540 --> 00:19:09.540 So I think this really illustrates my interpretation of the calculation problem because lots of 165 00:19:09.540 --> 00:19:16.380 times the emphasis is on not being able to allocate capital goods properly, but as I've 166 00:19:16.380 --> 00:19:22.740 tried to show, it is possible to allocate capital goods according to the central planner's 167 00:19:22.740 --> 00:19:28.940 There's goals if they solve the knowledge problem but there can be no intellectual division 168 00:19:28.940 --> 00:19:33.180 of labor under socialism. 169 00:19:33.180 --> 00:19:42.540 Now if you look at the examples that Mises gives of the calculation problem, they're 170 00:19:42.540 --> 00:19:50.740 all examples that try to abstract a part of the economy from the whole. 171 00:19:50.740 --> 00:19:57.740 He'll say, what if you try to build a house? Or what if the socialist planner wants to build a railroad? 172 00:19:57.740 --> 00:20:03.740 And the conclusion is always that this is impossible because there's no objective sense of profit and loss. 173 00:20:03.740 --> 00:20:07.740 There's no reason to believe that it's a good idea or a bad idea. 174 00:20:07.740 --> 00:20:13.740 But if you can think about everything happening in the entire economy all at once, 175 00:20:13.740 --> 00:20:16.740 then you are able to come up with an answer. 176 00:20:16.740 --> 00:20:32.060 Mises, because you can just compare the entire outcome from one to the other. 177 00:20:32.060 --> 00:20:41.300 And so, by conclusion, under socialism there can be no intellectual division of labor. 178 00:20:41.300 --> 00:20:44.980 And this is what is really impossible under socialism. 179 00:20:44.980 --> 00:20:53.980 It's not really the allocation of capital goods, but just the fact that there can be no intellectual division of labor. 180 00:20:53.980 --> 00:21:02.980 So here's a quote from Brian Kaplan, Why I'm Not an Austrian Economist. 181 00:21:02.980 --> 00:21:07.980 Does Trousseau's one-man socialism become impossible when Friday shows up? Hardly. 182 00:21:07.980 --> 00:21:09.980 What if a hundred people show up? One thousand. 183 00:21:09.980 --> 00:21:13.980 2000, Mises' distinction between a modern economy and proofs of it and why the economic 184 00:21:13.980 --> 00:21:19.340 calculation argument applies only to the former, again shows that Mises has underlying quantitative 185 00:21:19.340 --> 00:21:22.140 assumptions in spite of his strictures against them. 186 00:21:22.140 --> 00:21:28.860 So I think that my interpretation of the calculations and knowledge problems explains this apparent 187 00:21:28.860 --> 00:21:37.500 paradox in the interpretation of Mises, because it shows that if you have an economy in which 188 00:21:37.500 --> 00:21:42.220 you can encompass in which you can imagine and think about everything all at once 189 00:21:42.220 --> 00:21:48.460 and it won't fall apart, you won't produce things that people want 190 00:21:48.460 --> 00:21:52.080 but 191 00:21:52.080 --> 00:21:56.820 that's why the two-person socialist economy will work 192 00:21:56.820 --> 00:22:03.820 and the large economy won't. So thank you very much.