Books / Digital Text

I. Money

2. Subjectivism in International Economics: Why Absolute Purchasing Power Parity Does Not Hold by Simon Bilo

Austrian* economists have not ventured into the field of international economics very often and most of the exceptions wrote their work a long time ago. This is the case with the work on money and credit by Mises (1953 [1924], esp. pp. 170–86), Hayek’s discussion of monetary nationalism (1999 [1937], esp. pp. 61–73), Machlup’s (1939, 1940) and Haberler’s (1950) contributions, and Rothbard’s brief discussion in Man, Economy, and State (2004 [1962], esp. pp. 828ff.).

Of the few recent contributions made to this field, two authored by Salerno (1994a; 1994b) highlight the subjectivist perspective that Mises (1953 [1924]) holds about the determinants of the purchasing power of money in geographically separate locations. Physically identical goods in different locations are different economic goods even if we assume away all transportation costs. Because people often value separate economic goods differently, prices of physically identical goods in different locations might vary even in general equilibrium.

The insight that there can be an equilibrium with different prices between physically identical goods in different locations is important from the perspective of the absolute purchasing power parity theory, which is one of the models that tries to explain foreign exchange rates. This theory assumes the law of one price and implies that equilibrium exchange rate must equalize prices of physically identical goods across different currency areas when the prices of the goods are converted into same currency. Currently available data, however, bring this idea of absolute purchasing power parity into question: the general consensus is that in spite of high variability of foreign exchange rates, it takes a number of years at best before the exchange rate adjusts to a deviation from parity (Rogoff 1996; Taylor and Taylor 2004). It is this “purchasing power parity puzzle” (Rogoff 1996) that Mises’s subjectivist view on purchasing power of money can explain: if physically identical goods in different locations are different economic goods, it is not surprising that they have different price tags when the prices are expressed in the same currency and that absolute purchasing power parity does not hold. Yet, at the same time, there can still be a tendency toward equilibrium in the exchange rate between two currencies. The equilibrium exchange rate, however, does not reflect the purchasing power parity condition but rather the subjective valuations of goods in each currency area, given the prices of those goods in their respective domestic currencies.

In what follows, I develop the argument from the previous paragraph. I first review the insights of Mises and Salerno on the subjectivist theory of the purchasing power of money and then look at how these insights apply in the setting of two currency areas with a floating foreign exchange rate. In conclusion, I formulate the underlying subjectivist theory of foreign exchange rates.

Subjective Valuation Differentiates Purchasing Power of Money Across Space

In the section on “Alleged Local Differences in the Cost of Living,” Mises (1953, pp. 175–78) stresses the importance of the position of goods in space when considering the valuation of those goods and their relative prices. He illustrates how important the location of goods is by comparing the prices in Karlsbad, a desired spa destination, and prices in other cities. While the same type of good costs more in Karlsbad than in other cities, the price difference is justified because goods in Karlsbad are perceived as different types of goods. In other words, “[i]f [person] has to pay more in Karlsbad for the same quantity of satisfactions, this is due to the fact that by paying for them he is also paying the price of being able to enjoy them in the immediate neighborhood of the medicinal springs” (Mises 1953, pp. 176–77).

To generalize the previous example, one can say that the position of a good in space matters — geographic location is an important characteristic of an economic good that can change one’s perception of this good, and consequently its value and price. Physically identical goods in different locations can then be priced differently even in equilibrium (Mises 1953, pp. 177–78; Salerno 1994b, pp. 251–52).

Arbitrage Does Not Equalize Purchasing Power of Money Across Space

One can object that while the demand for goods might differ by location, the difference at least does not apply in the case of tradable goods, which can be easily transported from one place to another. The demand for apples in the city of Meadville in Northwestern Pennsylvania, for example, might be lower than the demand for apples in Manhattan, incentivizing suppliers to distribute apples accordingly and eventually equalize the prices of apples in both places. If the existing relative supply of apples in these two places results in lower relative price of apples in Meadville, this incentivizes entrepreneurs to ship apples from Meadville to Manhattan to equalize the profits from selling apples in the two different places. Assuming perfect competition and zero transportation costs, one might say that profits equalize when the price of apples in Meadville is the same as the price of apples in Manhattan.

However, since tradable goods are usually bundled with non-tradable complements as Rogoff (1996, pp. 649–50) and Taylor and Taylor (2004, pp. 136–37) briefly note, location also affects the prices of tradable goods. Shelf-space, for example, is one such non-tradable complement: returning to the apple parable, a sufficient lack of shelf-space in Manhattan may fail to incentivize shop-keepers to supply enough apples to equalize prices between Meadville and Manhattan. In this case, the opportunity cost of supplying so many apples is too high; Manhattan shop-keepers would rather use the scarce shelf-space to offer other products while keeping the price of apples relatively high.

To generalize the example, one can say that tradable goods often need to be bundled with non-tradable complements when sold in specific geographic locations. Since these complements might be subjectively valued and priced differently across locations, opportunities to arbitrage price differentials across space are limited. This limitation might then lead to price differentials between physically identical goods sold in different geographic locations.

Subjective Valuation Differentiates Purchasing Power of Money also Across Currency Areas

The conclusion that physically identical goods can vary in equilibrium prices between different locations also applies to the case of two separate currency areas. This application suggests that foreign exchange rates do not necessarily correspond to the absolute purchasing power parity of the respective currencies. To illustrate this point, I will use a modified version of the previous section’s apple parable.

Assume that there are only two places in the world: Manhattan and London. Each city has its own independent fiat currency so that people in Manhattan use the dollar ($) and people in London use the pound (₤). Let’s assume an equilibrium where an apple in Manhattan costs $6 and where a physically identical apple located in London sells for ₤2. Assuming away transportation costs, the absolute purchasing power parity theory says that the equilibrium exchange rate between dollars and pounds is $6 per ₤2, i.e., $3/₤1. If the foreign exchange rate was different, the purchasing power parity theory suggests that this would create a state of disequilibrium with associated arbitrage opportunities that buyers and sellers will exploit until the exchange rate $/₤ is equal to the ratio of the price of apple expressed in dollars over the price of apple expressed in pounds.

However, the subjectivist insight proposed by Mises (1953) and emphasized by Salerno (1994a; 1994b) suggests a very different conclusion about the equilibrium exchange rate. Following the example, even if $6 and ₤2 are the equilibrium prices of apple in Manhattan and London respectively, the two prices tell us little about the equilibrium foreign exchange rate between dollars and pounds. The difference in geographic location means that apple in Manhattan and apple in London represent two different economic goods. The difference means that while $6 is the price of an apple in Manhattan, we cannot necessarily infer from this that in equilibrium people are willing to pay the pound equivalent of $6 for an apple in London. People might be paying more or less for an apple in London than its dollar equivalent, depending both on the demand for apples in London and on the prices and subjective values of complementary non-tradable goods necessary to sell apples in London. Assuming that the equilibrium price of an apple in London is ₤2, this implies the exchange rate $/₤ can be below or above the absolute purchasing power parity of $3/₤1.

Purchasing power of money is therefore unequal across currency areas in the same way it is unequal across different geographic locations within the same currency area. Goods with identical physical characteristics but different locations are different economic goods (Salerno 1994a, p. 107). In equilibrium, such goods can have different prices when their respective prices are converted into the same currency unit. As a result, equilibrium foreign exchange rate does not have to equalize the prices of goods across currency areas and therefore does not have to adhere to the absolute purchasing power parity condition.

Foreign Currency is Valued Subjectively as a Means Toward Goods in Its Currency Area/p>

If absolute purchasing power parity is not the equilibrium condition for the foreign exchange rate between two currencies, what are the equilibrium conditions? It is important to realize in this regard that people demand money because it is medium of exchange (Mises 1953, pp. 30ff.) — a medium of directly purchasing goods in its corresponding currency area. Assuming that money does not have non-monetary uses, people value different currencies against each other depending on the economic goods they can procure with those respective currencies (Mises 1953, pp. 180–81).

The foreign exchange rate of a currency thus depends on the prices that people expect to pay for goods using the currency. If expected prices increase in one currency, demand for that currency drops at the foreign exchange market and its exchange rate becomes less favorable; if the expected prices decrease, the demand for the currency increases and its exchange rate becomes more favorable. In contrast to the absolute purchasing power parity theory, however, the relationship between the foreign exchange rate between two currencies and the prices of goods that people using each currency can buy is qualitative and does not follow a pre-determined mechanical formula. The numerical imprecision of the law explaining determinants of foreign exchange rates is a necessary consequence of the fact that most of the goods that people buy with each currency are different economic goods that people value subjectively. People’s subjective valuations therefore act as a filter for every price change of a good expressed in that currency: people ultimately decide to what extent the price change has an effect on their demand for the currency in question.

Conclusion: Subjectivism and International Economics

In his 1994a and 1994b articles, Salerno restored attention regarding Mises’s subjectivist approach to monetary theory and international economics. This approach helps us to understand why economists have been struggling to empirically confirm the absolute version of the purchasing power parity theory. They have been unsuccessful because the theory assumes the law of one price for goods that have identical physical characteristics but which differ in location. Because the difference in location means that these goods are in reality different economic goods, the law of one price does not have to hold and the absolute purchasing power parity can be violated even in equilibrium. The subjectivist approach to international economics thereby gives us yet another illustration of the importance of subjectivism in economics that was emphasized by Hayek (1952, p. 31).


Haberler, Gottfried. 1950. The Theory of International Trade. William Hodge & Company.

Hayek, Friedrich A. von. 1952. The Counter-Revolution of Science. Glencoe, Ill.: Free Press.

——. 1999 [1937]. “Monetary Nationalism and International Stability.” In Stephan Kresge, ed., The Collected Works of F. A. Hayek, Vol. 6: Good Money Part II: The Standard, pp. 37–100. London: University of Chicago Press and Rutledge.

Machlup, Fritz. 1939. “The Theory of Foreign Exchanges.” Economica, n.s. 6(24): 375–97.

Machlup, Fritz. 1940. “The Theory of Foreign Exchanges.” Economica, n.s. 7(25): 23–59.

Mises, Ludwig von. 1953 [1924]. The Theory of Money and Credit. New Haven, Conn.: Yale University Press.

Rogoff, Kenneth. 1996. “The Purchasing Power Parity Puzzle.” Journal of Economic Literature 34(2): 647–68.

Rothbard, Murray N. 2004 [1962]. Man, Economy, and State with Power and Market. Scholar’s Edition. Auburn, Ala.: Mises Institute.

Salerno, Joseph T. 1994a. “Ludwig von Mises’s Monetary Theory in Light of Modern Monetary Thought.” Review of Austrian Economics 8(1): 71–116.

——. 1994b. “International Monetary Theory.” In Peter Boettke, ed., The Elgar Companion to Austrian Economics, pp. 249–57. Aldershot, Hants, England and Brookfield, Vermont: Edward Elgar.

Taylor, Alan M., and Mark P. Taylor. 2004. “The Purchasing Power Parity Debate.” Journal of Economic Perspectives 18(4): 135–58.

  • *. Simon Bilo is assistant professor of economics at Allegheny College, Meadville, Pennsylvania. This paper is a revised version of selected sections of my 2006 M.A. thesis. I would like to thank Peter Boettke, Per Bylund, Gene Callahan, Jan Havel, Marek Hudík, Juraj Karpiš, Shruti Rajagopalan, Walter Stover, Lawrence White, and participants of the Graduate Student Paper Workshop at GMU for their valuable comments and suggestions during earlier drafts of this paper. A draft of the paper was also presented at the Austrian Scholars Conference in 2009. I gratefully acknowledge the financial help that I received from the Mercatus Center at George Mason University while working on this project. All the usual caveats apply. I have known Joseph Salerno for about ten years. These were ten formative years for me — I was an undergraduate student in Prague back then; now I am teaching economics myself. Salerno played an important role in this journey of mine: he was my adviser in the summer of 2005 at the Mises Institute, he kindly agreed to write letters of recommendation for me when I was applying for graduate school, and we would see each other when the two of us were attending the Colloquium on Market Institutions and Economic Processes at New York University.