Its Unassailable Essence and a Proposed Refinement.
Does the existence of Bitcoin, and other cryptocurrencies, invalidate Ludwig von Mises’s Regression Theorem? No, economic phenomena cannot invalidate economic theory, just as topographical measurements cannot invalidate geometry.
When economic phenomena and theory do not seem to agree, that can be an indication of inapplicability of the theory to the given phenomenon, just as a disagreement between triangle measurements and the Pythagorean Theorem may be due to the fact that the triangle being measured is not right, but slightly acute.
This is what some critics of Bitcoin think is happening in this case. They say that Bitcoin is not money, and therefore not the kind of phenomenon that the Regression Theorem refers to.
This position is untenable, because, as Mises himself stated clearly, the Regression Theorem concerns any kind of medium of exchange, and not just a general medium of exchange (money). So whether Bitcoin is money or not is irrelevant. It is certainly used as a medium of exchange by some people, even if it is not the general medium of exchange. And so, the Regression Theorem does indeed refer to Bitcoin.
Seeming disagreement between economic phenomena and theory can also be an indication of the misapprehension of the phenomena in question, just as disagreement between triangle measurements and the Pythagorean Theorem may be due to the fact that, although the triangle being measured is a right triangle, its sides were measured incorrectly; perhaps the measurer accidentally used the inches edge of his ruler for one side, while using the centimeters edge for the other sides.
This is what some proponents of Bitcoin think is happening in this case. They say that, contrary to common perception, Bitcoin does have commodity/use value, and not just monetary/exchange value, and therefore its existence is perfectly consistent with the Regression Theorem.
However, the issue is not whether Bitcoin has any use value at all, but whether its non-monetary use value is what first caused it to have any exchange value whatsoever. All of the sources of use value offered by Bitcoin proponents making this argument either (A) are not independent of Bitcoin’s prospective role as a medium of exchange, or (B) were not and could not have been broadly and highly valuable enough to alone give Bitcoin non-negligible liquidity.
More fundamentally, the issue is not whether Bitcoin’s rise as a medium of exchange really was independent any non-monetary use value, but whether it conceivably could have been so independent.
“…no good can be employed for the function of a medium of exchange which at the very beginning of its use for this purpose did not have exchange value on account of other employments. (…) It must happen this way. Nobody can ever succeed in construction a hypothetical case in which things were to occur in a different way.”
However, it seems to me that such a hypothetical case can be made. Whether or not it actually occurred, we can hypothesize that the mysterious Satoshi Nakamoto (Craig Wright?) created Bitcoin solely based on the expectation of it having future exchange value as a money. Whether or not it actually occurred, we can hypothesize the first person who sold him something for Bitcoins did so, because Nakamoto convinced him of its money-ish qualities and that therefore it would have future exchange value as a money. Once these assumptions are made, this gives us a medium of exchange whose regression never terminates in other employments than as a medium of exchange. Whether it happened that way or not, it is conceivable, and the hypothesis involves no inner contradictions.
This points at a third possible explanation for disagreement between economic theory and phenomena: perhaps the theory is indeed imperfect. Yet, any imperfection an economic theory has can only be due to imperfect reasoning. Imperfect reasoning can result in disagreement between theory and phenomena. And therefore, disagreeing phenomena may be an indication that the theory was constructed with imperfect reasoning. But disagreeing phenomena alone does not invalidate the theory. It only provides a prompt for the theorist to check his reasoning. It is then the discovery of imperfect reasoning alone that can invalidate an economic theorem.
Similarly, if a geometry student thought that, for a right triangle, the sum of the squares of the legs equalled the cube (and not the square) of the hypotenuse, it is not inconsistent measurements that would invalidate such a theorem. However, such measurements may prompt the geometry student to seek out the faults in his reasoning that resulted in such a theorem. It is only the identification of those faults that invalidates a false geometric theorem.
So, the existence of Bitcoin does not, and could never, invalidate the Regression Theorem, or any other economic theorem (the Law of Demand, the Law of Comparative Advantage, etc.). It can, however, prompt us to check our reasoning. And by thinking through imaginary constructions, we can realize that it is perfectly conceivable that a digital, cryptographic “coin” could be employed for the function of a medium of exchange for the first time without ever having an exchange value based on “commodity” employments. It is perfectly conceivable that any such medium-of-exchange employments may be based entirely on forecasted exchange value. And it is not necessary for a record of past valuations to exist as reference points in order that future exchange value can be forecasted. It is perfectly conceivable that the first person who traded a pizza for Bitcoin, for example, had some vague notion that Bitcoin might some day become a valuable monetary unit, just based on, say, Nakamoto’s arguments to that effect.
So, then, does pure reasoning (which may happen to be prompted by the existence of Bitcoin) invalidate the Regression Theorem? No. It does modify it. But the essential core of the Regression Theorem is totally true and indispensable to economic theory. Some proponents of Bitcoin think otherwise, but only because of a common misunderstanding of what the essence of the Regression Theorem is.
People put way too much emphasis on the way that Mises has his theorem’s regression terminate. But that is not what makes the Regression Theorem such an important advance for economics. After all, it is not called the Regression Terminus Theorem. The key element of the theorem is not the regression’s terminus, but the regression itself.
Mises explained in Human Action that:
…the demand for a medium of exchange is the composite of two partial demands: the demand displayed by the intention to use it in consumption and production and that displayed by the intention to use it as a medium of exchange. With regard to modern metallic money one speaks of the industrial demand and of the monetary demand. The value in exchange (purchasing power) of a medium of exchange is the resultant of the cumulative effect of both partial demands.
Now the extent of that part of the demand for a medium of exchange which is displayed on account of its service as a medium of exchange depends on its value in exchange. This fact raises difficulties which many economists considered insoluble so that they abstained from following farther along this line of reasoning. It is illogical, they said, to explain the purchasing power of money by reference to the demand for money, and the demand for money by reference to its purchasing power.
Since “the demand for a medium of exchange which is displayed on account of its service as a medium of exchange” is a function of its “service”/utility, and “purchasing power” is just another term for a money’s value in exchange, the problem boils down to the seemingly circular logic of “value comes from utility, which comes from value.”
This problem of circularity is why, before Mises’s great contribution in his 1912 Theory of Money and Credit, many theorists criticized the marginal utility theory of value, and the subjectivist economics that was based on it, as being discredited by its seemingly problematic application to money. Mises’s great contribution was to vindicate the “modern economics” that his Austrian predecessor Carl Menger co-founded, by breaking out of this circle with his regression. As Mises continued:
The difficulty is, however, merely apparent. The purchasing power which we explain by referring to the extent of specific demand is not the same purchasing power the height of which determines this specific demand. The problem is to conceive the determination of the purchasing power of the immediate future, of the impending moment. For the solution of this problem we refer to the purchasing power of the immediate past, of the moment just passed. These are two distinct magnitudes. It is erroneous to object to our theorem, which may be called the regression theorem, that it moves in a vicious circle.
The evolution and current state of the demand for and purchasing power of any widely used medium of exchange simply cannot be understood without the “temporal regression” essence of Mises’s Regression Theorem. Again, it is absolutely essential for sound, subjective-value-based economic theory.
Mises’s addition of the “commodity terminus” to the regression was merely a matter of tying up a loose end.
But, say the critics, this is tantamount to merely pushing back the problem. For now one must still explain the determination of yesterday’s purchasing power. If one explains this in the same way by referring to the purchasing power of the day before yesterday and so on, one slips into a regressus in infinitum. This reasoning, they assert, is certainly not a complete and logically satisfactory solution of the problem involved. What these critics fail to see is that the regression does not go back endlessly. It reaches a point at which the explanation is completed and no further question remains unanswered. If we trace the purchasing power of money back step by step, we finally arrive at the point at which the service of the good concerned as a medium of exchange begins. At this point yesterday’s exchange value is exclusively determined by the nonmonetary — industrial — demand which is displayed only by those who want to use this good for other employments than that of a medium of exchange.
This “commodity terminus” is perfectly conceivable and extremely likely in nearly all cases, but again, it is not the only terminus conceivable. Another conceivable terminus, especially in a world in which money is already a well-known phenomenon, is the initial instance of exchange value being entirely determined by monetary demand based purely on “hunch”-based forecasts of future purchasing power. Once a record of purchasing power is established, past purchasing power must be taken into account when explaining present purchasing power. But there needn’t necessarily be such a record for the first flickering of purchasing power to occur.
Ludwig von Mises’s Regression Theorem was one of the most brilliant and crucial advances in the history of monetary economics. It is only its merely auxiliary proposition concerning the regression’s terminus that should be modified in light of the conceivability of digital cryptocurrencies. The Regression Theorem’s essential proposition stands as indispensable and unassailably true today as it did in 1912, and as it will in 3012.
Also published at Medium.com: