Solving the Wallis–North paradox in transaction cost measurement

This note contributes to solving the Wallis–North paradox in transaction cost measurement. By distinguishing latent transaction costs from actual transaction costs, the note showed that when the latent transaction costs were prohibitively high, no transaction and actual transaction costs would happen and thus the share of transaction sectors should be zero; nevertheless, with latent transaction costs declining, transaction sectors and actual transaction costs will increase. Therefore, a higher share of transaction sectors in an economy reveals that the economy actually has a lower level of (latent) transaction costs. The Wallis–North approach is still workable for the transaction cost measurement but with an inverted interpretation.