WEBVTT NOTE The Regression Theorem as Conjectural History 1 00:00:00.000 --> 00:00:13.800 In assessing the importance of the theory of money and credit, I'd like to go back at least briefly giving a quick overview of why it really was a turning point. 2 00:00:13.800 --> 00:00:23.600 If we say that Menger, Boehm-Bawerk and Wieser were the first generation of Austrian economists, 3 00:00:23.600 --> 00:00:34.760 We date Mises as being the second generation, with Hayek and Robbins and Röpke being the third, 4 00:00:34.760 --> 00:00:44.360 and then with Haslett and Rothbard and those associated with Mises' seminar as being the fourth generation. 5 00:00:44.360 --> 00:00:53.440 I think it's fair then to say what was unique about Mises that made him an important transitional figure 6 00:00:53.440 --> 00:00:59.960 I think almost as a stand-alone figure for that second generation. 7 00:00:59.960 --> 00:01:07.840 And I would like to read very briefly from a footnote that appears in Hayek's book, 8 00:01:07.840 --> 00:01:12.200 The Counter-Revolution of Science, which was published in 1952. 9 00:01:12.200 --> 00:01:18.200 And in the footnote, it's on page 52 of the reprint by the Liberty Fund, 10 00:01:18.200 --> 00:01:23.040 he makes an important, very important observation. 11 00:01:23.040 --> 00:01:26.120 And I will quote it verbatim. 12 00:01:26.120 --> 00:01:29.480 This is a development which has probably been carried out 13 00:01:29.480 --> 00:01:32.800 most consistently by Ludwig von Mises. 14 00:01:32.800 --> 00:01:37.440 And I believe most peculiarities of his views, 15 00:01:37.440 --> 00:01:41.240 which at first strike many readers as strange 16 00:01:41.240 --> 00:01:45.160 and unacceptable, trace to the fact 17 00:01:45.160 --> 00:01:48.360 that in the consistent development 18 00:01:48.360 --> 00:01:51.120 of the subjectivist approach, 19 00:01:51.120 --> 00:01:57.520 he has for a long time moved ahead of his contemporaries. 20 00:01:57.520 --> 00:02:07.120 Now the theory of money and credit was clearly and powerfully a defense of subjective value theory 21 00:02:07.120 --> 00:02:16.120 and the application of subjective value theory to the central institution of the modern market economy, 22 00:02:16.120 --> 00:02:18.320 which is the money system. 23 00:02:18.320 --> 00:02:33.200 And Mises was rigorous in applying subjectivism to that area that was so important in the development of his own theory of the division of labor and the rise of the modern market economy. 24 00:02:33.200 --> 00:02:37.880 And you have to keep that in the back of your mind whenever you read The Theory of Money and Credit. 25 00:02:37.880 --> 00:02:47.880 Think to yourself, subjective theory of value, subjective theory of value, he is reconstructing monetary theory in terms of subjectivism. 26 00:02:47.880 --> 00:02:55.880 The book was published, as you know, in 1912. 27 00:02:55.880 --> 00:03:04.880 Twenty years earlier, Menger wrote his final academic essay, that is in 1892, 28 00:03:04.880 --> 00:03:08.640 and it dealt with the theory of money. 29 00:03:08.640 --> 00:03:13.640 It was in a limited sense, 30 00:03:14.280 --> 00:03:18.120 it was an attempt to take subjectivism 31 00:03:18.120 --> 00:03:20.880 and apply it to monetary theory. 32 00:03:20.880 --> 00:03:25.640 But Mises took up that concept 33 00:03:25.640 --> 00:03:28.480 and applied it much more rigorously. 34 00:03:28.480 --> 00:03:33.240 That is, he was extending Menger's original insight 35 00:03:33.240 --> 00:03:40.200 on Subjectivism and extending it beyond that one essay that Menger wrote in 92. 36 00:03:40.200 --> 00:03:48.400 Menger's essay began with a legitimate question, that is, the question of the origin of money. 37 00:03:48.400 --> 00:03:56.020 And he made the observation, which we find probably equally bizarre today, he made the 38 00:03:56.020 --> 00:04:03.300 The observation that why is it that pieces of metal should be at the central part of 39 00:04:03.300 --> 00:04:05.220 the division of labor economy? 40 00:04:05.220 --> 00:04:12.900 Why is it that throughout the world, and it was a better world, that pieces of metal, 41 00:04:12.900 --> 00:04:20.180 gold and silver, would serve internationally across wide geographical areas as the means 42 00:04:20.180 --> 00:04:21.180 of exchange? 43 00:04:21.180 --> 00:04:26.100 said there's nothing innate about that, there's nothing automatic about that, and then he 44 00:04:26.100 --> 00:04:33.100 went back and he said if we look historically, most analysts have said that it is the declaration 45 00:04:33.100 --> 00:04:41.340 by the state that a particular money metal would be used widely in society, and that 46 00:04:41.340 --> 00:04:48.840 it is state sovereignty that is the source of monetary value. Now he opposed that view 47 00:04:48.840 --> 00:04:50.560 You, of course. 48 00:04:50.560 --> 00:04:57.280 But if you look through the history of monetary theory, he took it back to Aristotle at least, 49 00:04:57.280 --> 00:05:01.680 and that state sovereignty is the center of the theory of money. 50 00:05:01.680 --> 00:05:03.960 Well that certainly has not changed in our day. 51 00:05:03.960 --> 00:05:09.640 In fact, it is much more believed today than it was in Menger's day. 52 00:05:09.640 --> 00:05:11.920 So he said, why should this be the case? 53 00:05:11.920 --> 00:05:12.920 And then he made an argument. 54 00:05:12.920 --> 00:05:18.440 It was a legitimate argument, but it had a tendency to backfire. 55 00:05:18.440 --> 00:05:19.960 and that's why I want to mention it. 56 00:05:19.960 --> 00:05:21.800 He said, if we look at historical records, 57 00:05:21.800 --> 00:05:26.800 we cannot find cases where declarations of state sovereignty 58 00:05:27.480 --> 00:05:30.400 fiat money, in other words, fiat declarations, 59 00:05:30.400 --> 00:05:34.460 speaking value into existence, where that ever took place. 60 00:05:34.460 --> 00:05:37.440 There are no records of any such event in society. 61 00:05:37.440 --> 00:05:39.780 So he said, if this was the basis of the origin 62 00:05:39.780 --> 00:05:41.800 of monetary theory, it would have left some records 63 00:05:41.800 --> 00:05:43.920 and there are no such records. 64 00:05:43.920 --> 00:05:48.160 So that, of course, led him to a kind of an Austrian view 65 00:05:48.160 --> 00:06:15.160 There is no center of population which has not, in the very beginnings of civilization, come keenly to desire and eagerly to covet the precious metals. 66 00:06:15.160 --> 00:06:36.160 in primitive times for their utility and particular beauty as in themselves ornamental, subsequently as choice materials for plastic and architectural decoration and especially for ornaments and vessels of every kind. 67 00:06:36.160 --> 00:06:49.160 One sentence, taking it back to the value of the metals for non-exchange purposes, what we might call use in kind. 68 00:06:49.160 --> 00:06:53.160 He said that's where the origin of the metals came from. 69 00:06:53.160 --> 00:06:55.160 Now just one sentence. 70 00:06:55.160 --> 00:07:04.160 But Mises took that, that one sentence insight and developed what we call the regression theorem. 71 00:07:04.160 --> 00:07:17.160 Now why was he so concerned about that? Because he opposed the concept of the sovereignty of the state as being the origin of any value and certainly not a monetary value. 72 00:07:17.160 --> 00:07:30.160 So if it was not the declaration of the state ratifying a particular money metal, if that was not the basis of the origin of money, precious metal in exchange, then what was it? 73 00:07:30.160 --> 00:07:31.160 What was it? 74 00:07:31.160 --> 00:07:42.000 Well, he took it back to where Menger had begun it in the area of life outside of exchange, 75 00:07:42.000 --> 00:07:46.600 in the areas of life of ornaments, and then of course you go much beyond it, especially 76 00:07:46.600 --> 00:07:49.140 in religious symbolism. 77 00:07:49.140 --> 00:07:53.880 Silver and gold have been widely used for as long as we have records. 78 00:07:53.880 --> 00:08:01.920 So that the metals were desirable for uses other than monetary exchange. 79 00:08:01.920 --> 00:08:10.960 And only through practice, in a sense, did the metals begin to perform as the means of 80 00:08:10.960 --> 00:08:11.960 exchange. 81 00:08:11.960 --> 00:08:20.100 Mises used the same reason that Menger used for the origin of the money metals as money, 82 00:08:20.100 --> 00:08:24.820 that they were the most exchangeable or saleable commodities. 83 00:08:24.820 --> 00:08:29.820 Mises defined money as the most marketable commodity. 84 00:08:31.080 --> 00:08:34.160 That insight was originally Menger's. 85 00:08:34.160 --> 00:08:39.160 So, because people imputed value to the metals 86 00:08:40.980 --> 00:08:45.980 in terms of exchange, the metals became used for money. 87 00:08:46.340 --> 00:08:47.840 Now, why did they do that? 88 00:08:47.840 --> 00:08:58.840 On the one hand, there was a future orientation, entrepreneurial saying, probably these metals will be usable in exchange tomorrow and the next day and next week and a hundred years from now. 89 00:08:58.840 --> 00:09:08.340 Why did they believe that? Because they could look behind them and say, well, these metals have been used yesterday and the day before and ten years and a hundred years back, 90 00:09:08.340 --> 00:09:22.340 So we can make a deduction based on history of exchange that the metals will be useful in the future and therefore, as the phrase we really should not use as Austrians but always is used, they become a store of value. 91 00:09:22.340 --> 00:09:29.140 Now what we really ought to say is that they are a valuable thing to store, which they are. 92 00:09:29.140 --> 00:09:39.780 Now the problem with this observation and the use that Mises put Menger's observation for, 93 00:09:39.780 --> 00:09:46.620 the problem is there's no historical evidence because of the nature of the evidence. 94 00:09:46.620 --> 00:09:51.640 That is you didn't have the science of accounting before you had money. 95 00:09:51.640 --> 00:09:56.800 Now there's no question that if you look into the ancient records you find a great deal 96 00:09:56.800 --> 00:10:12.800 The deal of accounting material, in fact, as I've said, it's almost mind-numbing, the amount of documentation that survived through the ages on this account versus that account interests all the aspects of money. 97 00:10:12.800 --> 00:10:18.800 Most historical records are tied to monetary transactions of one kind or another. 98 00:10:18.800 --> 00:10:27.600 Well, the question is how they leave those kinds of records in the transition period before money became the means of exchange and records became possible. 99 00:10:27.600 --> 00:10:30.800 The answer is we don't have the records. 100 00:10:30.800 --> 00:10:44.400 Now, the problem then is, does the critique of Menger against those who said the sovereignty of the state is the origin of money, but there's no record, said Menger. 101 00:10:44.400 --> 00:10:50.880 Does that then, in effect, torpedo Mises's regression theorem? 102 00:10:50.880 --> 00:10:58.360 Because there's not a wide body of historical materials to support his theory either. 103 00:10:58.360 --> 00:11:00.520 So how do we deal with it? 104 00:11:00.520 --> 00:11:03.960 It's a problem of epistemology, ultimately. 105 00:11:03.960 --> 00:11:07.340 Epistemology being what does man know and how can he know it? 106 00:11:07.340 --> 00:11:15.000 How do we know that money really did have its origin in this transfer of subjective 107 00:11:15.000 --> 00:11:23.560 value from ornaments and religious objects over to monetary exchange? 108 00:11:23.560 --> 00:11:25.580 We don't have the records for it. 109 00:11:25.580 --> 00:11:29.260 Mises never implied that there were records for it. 110 00:11:29.260 --> 00:11:36.540 And yet the logic of it, according to Austrian categories, is that this must have been how 111 00:11:36.540 --> 00:11:38.780 money came into existence. 112 00:11:38.780 --> 00:11:42.700 Because value had to originate in the individual 113 00:11:42.700 --> 00:11:45.600 who made subjective valuations, 114 00:11:45.600 --> 00:11:48.040 but it couldn't have been a subjective valuation of money 115 00:11:48.040 --> 00:11:50.760 because the metals didn't function as money. 116 00:11:51.660 --> 00:11:55.000 Now, we Missesians, we look at the regression theorem 117 00:11:55.000 --> 00:11:57.380 and says, well, that's just common sense. 118 00:11:57.380 --> 00:12:00.300 But yet in the history of discussions of money, 119 00:12:00.300 --> 00:12:01.420 it's not common sense. 120 00:12:01.420 --> 00:12:04.320 That's why the theory of money and credit 121 00:12:04.320 --> 00:12:07.080 and the regression theorem was a breakthrough 122 00:12:07.080 --> 00:12:09.640 because it was not common sense. 123 00:12:09.640 --> 00:12:12.120 Mises went instead of saying that 124 00:12:12.120 --> 00:12:14.680 the state was the origin of value of money, 125 00:12:14.680 --> 00:12:17.640 he said individual imputation was the source 126 00:12:17.640 --> 00:12:20.820 and that was a radical subjectivist position 127 00:12:20.820 --> 00:12:23.860 and it was a major breakthrough in the theory of money. 128 00:12:24.880 --> 00:12:26.640 So how are we gonna sort this out? 129 00:12:27.560 --> 00:12:30.960 If you go to Mises, especially in Human Action, 130 00:12:30.960 --> 00:12:39.200 And now that we can do searches digitally, look up the phrase, apodictic. 131 00:12:39.200 --> 00:12:42.960 Oh, he loved to use that one. Apodictic certainty. 132 00:12:42.960 --> 00:12:50.440 That probably angered the methodologists in the rival fields more than just about anything else the man ever said. 133 00:12:50.440 --> 00:12:58.920 Apodictic certainty. It sends them into a kind of near paralysis of anger when they see a phrase like that. 134 00:12:58.920 --> 00:13:12.400 What did he mean? He meant, because he was a deductivist, he meant something like, as the price increases, less of it will be demanded. 135 00:13:12.400 --> 00:13:20.280 Well, of course, your standard economist says, yes, yes, yes, we know that, but Mises would say that's a statement of apodictic certainty. 136 00:13:20.280 --> 00:13:23.960 It's a deduction from the nature of human action. 137 00:13:23.960 --> 00:13:35.460 And Mises, in a sense, made or broke his reputation in the academic world with that very approach, methodologically, to studying economics. 138 00:13:35.460 --> 00:13:39.460 And it was all about apodictic certainty. 139 00:13:39.460 --> 00:13:44.960 Then there's historical evidence, and there is historical evidence, and he talked about that, 140 00:13:44.960 --> 00:13:51.200 but he didn't give the same weight of importance to historical evidence that he did to the 141 00:13:51.200 --> 00:13:56.020 logic of economics as a deductivist system. 142 00:13:56.020 --> 00:14:04.200 So if it's not historical evidence with footnotes of this and that society all going from religious 143 00:14:04.200 --> 00:14:12.200 objects over to the use of gold and silver in exchange, if it's not that kind of history 144 00:14:12.200 --> 00:14:18.200 and clearly it isn't, but if it's also not apodictic certainty in terms of an extension 145 00:14:18.200 --> 00:14:23.200 of the categories of human action, then what is it? 146 00:14:23.200 --> 00:14:25.480 What do we call this thing? 147 00:14:25.480 --> 00:14:34.400 In the late 1960s, I studied sociology with Robert Nisbett and I was very much influenced 148 00:14:34.400 --> 00:14:41.440 by his writing, no question about it, and especially by his book, Social Change in History. 149 00:14:41.440 --> 00:14:48.520 In that book, he made an important observation about exactly the kind of knowledge that we're 150 00:14:48.520 --> 00:14:51.560 talking about when we talk about the regression theorem. 151 00:14:51.560 --> 00:15:00.040 Here's what he said, the emphasis is not upon the past as conceived as a genealogy 152 00:15:00.040 --> 00:15:07.800 of happenings and persons, but upon more or less timeless sequences of emerging changes. 153 00:15:07.800 --> 00:15:15.800 If event is the key to the historiographic perspective, change is the key concept in 154 00:15:15.800 --> 00:15:19.480 the developmental perspective. 155 00:15:19.480 --> 00:15:25.240 Time in the very broad sense matters, of course, but the developmentalist, whether social or 156 00:15:25.240 --> 00:15:32.600 biological, is far more interested in arriving at a correct before and after relationships 157 00:15:32.600 --> 00:15:41.400 in his changes and types than in the probably futile search for dates as to when exactly 158 00:15:41.400 --> 00:15:44.640 a certain change occurred. 159 00:15:44.640 --> 00:15:51.700 He talks about developmentalism or evolutionism or social evolutionism as a separate category 160 00:15:51.700 --> 00:15:56.080 of knowledge and he said we go back to it over and over and social theory goes back 161 00:15:56.080 --> 00:15:57.440 to it over and over. 162 00:15:57.440 --> 00:16:02.440 It's not based on all the footnotes of this event and that event and somebody exchanged 163 00:16:02.440 --> 00:16:08.600 It's not that kind of knowledge, but it's also not apodictic certainty. 164 00:16:08.600 --> 00:16:10.600 It's something in the middle. 165 00:16:10.600 --> 00:16:16.840 And in fact, most of what we do as economists and as historians, most of what we know about 166 00:16:16.840 --> 00:16:25.580 the past is much more developmentalist than it is the pile of footnotes or the brilliant 167 00:16:25.580 --> 00:16:28.480 logic of analysis. 168 00:16:28.480 --> 00:16:31.080 It's somewhere in between. 169 00:16:31.080 --> 00:16:37.080 And I think if I have anything to say that's important, and it's not radically important, 170 00:16:37.080 --> 00:16:43.560 it's just to say that this is back to a problem of epistemology, and we Austrians are always 171 00:16:43.560 --> 00:16:49.080 caught up in these questions of epistemology, really more than most of the other schools 172 00:16:49.080 --> 00:16:54.720 of thought, and that's because of what Mises did beginning in The Theory of Money and Credit. 173 00:16:54.720 --> 00:17:01.060 In applying radically the deductivist approach, based on logical categories, which you might 174 00:17:01.060 --> 00:17:05.780 might call the Parmenidean model versus the Heraclitus model 175 00:17:05.780 --> 00:17:09.780 of one thing after another, you can't put your foot 176 00:17:09.780 --> 00:17:12.060 into the same river twice and so forth. 177 00:17:12.060 --> 00:17:16.380 The sweep of history and the logical categories 178 00:17:16.380 --> 00:17:18.440 we find in our work as economists 179 00:17:18.440 --> 00:17:20.740 and just about every, even entrepreneurs, 180 00:17:20.740 --> 00:17:22.960 we're dealing in that middle area, 181 00:17:22.960 --> 00:17:25.360 but we're not dealing with the historical footnotes, 182 00:17:25.360 --> 00:17:27.800 we're not dealing with apodictic certainty, 183 00:17:27.800 --> 00:17:29.200 we're dealing with something much more 184 00:17:29.200 --> 00:17:30.700 like the regression analysis. 185 00:17:30.700 --> 00:17:36.700 That is, what we think must have happened, but we can't prove it from the historical record. 186 00:17:36.700 --> 00:17:43.700 But you have to come to this conclusion if you don't say that the state, by fiat diktat, 187 00:17:43.700 --> 00:17:53.700 established value of money as Austrians, we are, in a sense, stuck with some version of the regression analysis. 188 00:17:53.700 --> 00:17:57.200 It's not that I'm opposed to it. I think it's logical, but we've got to know what we're doing. 189 00:17:57.200 --> 00:18:02.440 It's not based on historical documentation, it's not also based on apodictic certainty. 190 00:18:02.440 --> 00:18:08.680 It's that area of epistemology of applied economics in between the two that we come 191 00:18:08.680 --> 00:18:11.920 to the concept of the regression theorem.