Using Engel Curves to Estimate Purchasing Power Parity. A Case Study of the Computation of the Exchange Rate between the Norwegian krone and the U.S. dollar
Standard practice of estimating purchasing power parities (PPP) involves using prices, in domestic currencies, of a common basket of goods and services, then calculating the price-equalizing exchange rate. In this article, I substitute observed consumer behavior for price data. On the assumption that an Engel curve for food reflects material standard of living, I estimate Engel curves for food for the United States and Norway. This allows us to calculate the exchange rate required for re-aligning the two curves, i.e. the incomes needed in the two countries to purchase the same standard of living. Since different relative prices or preferences for food can affect the position and slope of the curves, I also estimate the Engel curves of non-food, for which the effect is opposite. Not only does this provide a band of upper and lower bounds of PPP, it also improves upon the assumption of preference homogeneity underlying conventional PPP-computations. Using Consumer Expenditure (CES) data for 2001, I obtain estimated PPP-levels for the rate of the Norwegian krone (NOK) versus the U.S. dollar (USD) in the 5.38-7.90 range. The average rate 1977-2007 was 6.81 NOK per USD. The conventional estimates of PPP from the World Bank and OECD are 8.84 and 9.18 NOK per USD, respectively.
|Date of creation:||Mar 2009|
|Date of revision:|
|Contact details of provider:|| Postal: P.O.Box 8131 Dep, N-0033 Oslo, Norway|
Phone: (+47) 21 09 00 00
Fax: (+47) 21 09 49 73
Web page: http://www.ssb.no/en/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:ssb:dispap:580. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (J Bruusgaard)
If references are entirely missing, you can add them using this form.