Relative purchasing power parity (PPP) holds for pure price inflations, which affect prices of all goods and services by the same proportion, while leaving relative prices unchanged. Pure price inflations also affect nominal returns of all traded financial assets by exactly the same amount. Recognizing that relative PPP may not hold for the official inflation data constructed from commodity price indices because of relative price changes and other frictions that cause prices to be "sticky," we provide a novel method for extracting a proxy for realized pure price inflation from stock returns. We find strong support for relative PPP in the short run using the extracted inflation measures.
Extracting Inflation from Stock Returns to Test Purchasing Power ParityPurchasing power parity (PPP) is the simple proposition that prices in different countries should be equal if they are converted to the same currency. The absolute version of PPP is based on the law of one price, which maintains that arbitrage should tend to equilibrate prices of the same good at different locations. If the composition of the basket of goods used for constructing price indices is identical across countries, PPP trivially follows from the law of one price.However, frictions to goods arbitrage such as transportation costs and other impediments to trade, the extreme being non-tradable goods such as land, inhibit cross-country price equalization. Even with such frictions, the relative version of PPP, which maintains that the change in price levels across countries should be the same after adjusting for the change in the exchange rate, may still hold if relative price changes across countries are identical. For instance, a pure money shock will change nominal prices of all goods, services and assets by exactly the same amount so that relative prices among different assets will remain constant.In this strict pure price inflation case, the relative version of PPP will hold.Although simple, the PPP hypothesis has defied empirical confirmation for decades.There seems to be little agreement about why it fails so spectacularly when taken to data.
1The absolute version of PPP may fail because of the above mentioned frictions to goods arbitrage, and is not the focus of this paper. Our focus instead is on the failure of relative PPP.1 Rather than provide a long list of relevant references here, we point the reader to an excellent survey by Rogoff (1996).
1The failure of relative PPP in its most basic form can be described as follows. If relative PPP holds, then changes in the exchange rate must equal the concurrent inflation differential between two countries. Empirically, the two are at most weakly correlated (Rogoff, 1996). Furthermore, changes in exchange rates are extremely volatile, with a yearly standard deviation typically on the order of 12-13% for developed countries, while inflation differentials have yearly standard deviations less than 1% (see Rogoff, 1998). This empirical regularity has led scholars to wonder if exchange rate movements ...