Tradability is the property of a good or service that can be sold in another location distant from where it was produced. A good that is not tradable is called non-tradable. Different goods have differing levels of tradability: the higher the cost of transportation and the shorter the shelf life, the less tradable a good is. Prepared food, for example, is not generally considered a tradable good; it will be sold in the city in which it is produced and does not directly compete with other cities' prepared foods. Some non-commodities and services such as haircuts and massages are also obviously non-tradable. However, in recent years even pure services such as education can be regarded as tradable due to advancements in information and communications technology. ^[1]

Tradability is not an inherent binary attribute of any commodity or service; rather, it is more of a gradient determined by the interplay of certain physical, economic, and technical aspects. The classic criterion is the cost of shipping a commodity from the location of its production to a remote destination: the lower the transportation cost in comparison to the value of the commodity, the more tradable it will be; and the higher, in comparison to the value, the more non-tradable it will be ^[2]. The traditional approach based on the criterion of transport costs constitutes the foundation of standard analyses of non-traded commodities in international trade theory ^[2], ^[3].

The other factor is perishability. Any good or service that must be consumed at or very close to the time and place of its production – prepared food, personal services like hair cutting, and a great deal of in-person care services – does not possess any life span during which it can be moved to the buyer, and hence they belong to one of the most un-tradable categories irrespective of the shipping cost ^[2].

The third barrier pertains to regulations and policies. In some cases, tariffs, quotas, licensing laws, and other trading constraints may make it impossible to trade in a good which could have otherwise been easily tradable because such constraints make trading unprofitable ^[4].

Fourthly, the more recent one relates to developments in information and communications technology (ICT). ICT developments have broadened the range of tradable services to include those that were previously thought to be deliverable only in the form of close contact between the supplier and the consumer, such as education services, customer services and some kinds of professional services, among others ^[5], ^[6].

It is these factors that give rise to the distinction between tradables and non-tradables in open-economy macroeconomic analysis. It is this distinction that is key to the real exchange rate literature and the Balassa-Samuelson model, where cross-country differences in price levels can be explained by differences in productivity that occur in the tradables sector only (see § Price equalisation) ^[2], ^[3].

Perfectly tradable goods, like shares of stock, are subject to the law of one price: they should cost the same amount wherever they are bought. This law requires an efficient market. Any discrepancy that may exist in pricing perfectly tradable goods because of foreign currency price movements, for instance, is called an arbitrage opportunity. Goods that cannot be costlessly traded are not subject to this law.

Less than perfectly tradable goods are subject to distortions such as the Penn effect, for example, a lowering of prices in less wealthy place. Perfectly non-tradable goods are not subject to any leveling of price, thus the disparity between similar parcels of real estate in different locations.

There should be no distortions in purchasing power parity for perfectly tradable goods. The differences between it and other methods are the result of non-tradable goods and the above-mentioned Penn effect.