price surveys, meaning that PPP rates are unlikely to be consistent
over time or between different estimates (Callen 2007).
The so-called “substitution bias” is another weakness of PPP
exchange rates. This refers to the practice of assigning U.S. prices to
services consumed by people in developing countries. In reality,
however, U.S. prices for services tend to be much higher than those in
developing countries, and PPP-derived income estimates are likely to
be inconsistent with actual consumption structures and result in
artificial substitution (Dowrick and Akmal 2005). Similar to this is the
fact that it is unrealistic to compare countries with very different
consumption patterns.
A further drawback to using PPPs is contrasting results. While there
are three available series of PPP-adjusted GDP data—Maddison, Penn
World Table and World Bank—all of which are based on the PPP
rates produced by the ICP, comparing these different sources
produces significant variations across countries. This means that PPP
income estimates will vary according to the data source selected
(Sutcliffe 2003).
Advantages of PPP: Many argue that PPP rates are better than market
rates when comparing GDP across countries because PPP attempts to
measure this value at a common set of prices. In particular, the
exchange rate measure implies that all national output is sold on world
markets and that all national consumption is imported—a very
unrealistic assumption often referred to as the “traded sector bias.”
Since non-traded goods and services tend to be cheaper in low-income
countries when compared to higher-income countries, any analysis
that fails to take these price differences into account will
underestimate the purchasing power of consumers in developing
countries and, consequently, their overall welfare or income share.
PPP exchange rates further have the advantage of being relatively
stable over time whereas market rates are more volatile.
Does it make a difference? The per capita income gap between the
richest and poorest global population quintiles—as well as individual
countries—is reduced under PPP exchange rates according to our
estimates, a finding that reflects the well-known fact that PPP
exchange rates are higher than market ones. Some countries also move
up or down the income scale depending on the metric used.
Irrespective of method, however, income disparities remain
exceptionally high.