Purchasing Power Parity
The exchange rate prices imply, and the one inflation predicts — against the market rate.
The rate that would make both prices equal
against USD on this basket
1 means parity holds
At the market rate, the GBP 3.49 abroad converts to USD 4.4177 — cheaper than the 5.69 at home. Parity would need 0.6134 GBP per USD; the market gives 0.79, so GBP is undervalued by 22.36% — equivalently USD is overvalued by 28.8%.
Purchasing power parity says the same good should cost the same everywhere once converted. Divide the two prices to get the implied rate and compare it with the market. A basket at 5.69 at home and 3.49 abroad implies 0.6134, so a market rate of 0.79 leaves the foreign currency 22.4% undervalued.
One good, two currencies
The logic starts from arbitrage. If an identical, easily traded good were cheaper abroad, buyers would shift there, bidding up its price and the currency needed to buy it, until the gap closed. The exchange rate at which the gap is exactly zero is the implied PPP rate: the foreign price divided by the home price.
Comparing that with the market rate gives the valuation. Where parity would need only 0.6134 units of the foreign currency per unit of home currency but the market charges 0.79, the foreign currency is trading below what its purchasing power warrants — it is undervalued, by 22.4%. The same gap read the other way makes the home currency overvalued by 28.8%; the two figures differ because a percentage is not symmetric under inversion.
Relative PPP asks a different question
Absolute PPP is about the level of the exchange rate and rarely holds. Relative PPP is about its change, and holds up much better: the currency of the higher-inflation country should depreciate by roughly the inflation gap. That version survives the objections about baskets and local costs, because those distortions largely cancel when you look at changes rather than levels.
rates quoted as foreign currency per unit of home currency
- 1 Price the same thing in both places. A basket costing 5.69 at home and 3.49 abroad, each in its own currency.
- 2 Divide to get the implied rate. 3.49 ÷ 5.69 = 0.6134 foreign units per home unit — the rate that would equalise the two prices.
- 3 Compare with the market. The market gives 0.79, which is higher, so the foreign currency is cheaper than parity says it should be.
- 4 Turn the gap into a percentage. 0.6134 ÷ 0.79 − 1 = −22.4%, so the foreign currency is undervalued by 22.4%.
- 5 Sanity-check by converting. The 3.49 abroad converts to 4.42 at home against 5.69 there — cheaper abroad, which is what undervaluation means in practice.
Where the inflation shortcut breaks down
Expected change in the home currency over one year, exact against the subtract-the-rates shortcut.
| Home inflation | Foreign inflation | Exact change | Shortcut | Gap |
|---|---|---|---|---|
| 2% | 2% | 0.00% | 0.00% | 0.00 pp |
| 3% | 5% | +1.94% | +2.00% | 0.06 pp |
| 10% | 2% | −7.27% | −8.00% | 0.73 pp |
| 60% | 3% | −35.63% | −57.00% | 21.37 pp |
| 100% | 5% | −47.50% | −95.00% | 47.50 pp |
Why parity fails in practice
Most of what you buy never crosses a border. A haircut, a bus fare and a restaurant meal are produced and consumed locally, so no arbitrage forces their prices into line — and they make up a large share of any real basket. This produces the Balassa–Samuelson effect: rich countries have higher productivity in traded goods, which pulls up wages economy-wide, which pushes up the price of non-traded services. Poorer countries therefore look systematically undervalued on PPP, not because their currencies are mispriced but because their haircuts are genuinely cheaper.
Tariffs, transport, taxes and brand pricing add more wedges, and none of them are errors to be corrected. This is why PPP comparisons of currencies are better read as a rough gauge of whether something is unusually far from parity than as a forecast, and why the International Monetary Fund and the World Bank use PPP conversion rather than market rates when comparing living standards across countries.
The table above shows a separate trap, in the arithmetic rather than the economics. The classroom shortcut of subtracting inflation rates comes from an approximation that drops a cross term, and that term is negligible at low inflation and enormous at high inflation — off by half a percentage point at 10% inflation, and by 47 percentage points at 100%. Where inflation is high, use the exact ratio (1 + π_foreign) ÷ (1 + π_home), which is what this page computes.