## An Unbiased Appraisal of Purchasing Power Parity

_IMF Working Papers, December 1, 2001_

## Source details

**Canonical URL:** [An Unbiased Appraisal of Purchasing Power Parity](https://www.imf.org/en/publications/wp/issues/2016/12/30/an-unbiased-appraisal-of-purchasing-power-parity-15512)

## Other formats

- [Markdown version](/en/publications/wp/issues/2016/12/30/an-unbiased-appraisal-of-purchasing-power-parity-15512/index.md)
- [Structured JSON version](/en/publications/wp/issues/2016/12/30/an-unbiased-appraisal-of-purchasing-power-parity-15512/index.json)
- [Bundle manifest](/en/publications/wp/issues/2016/12/30/an-unbiased-appraisal-of-purchasing-power-parity-15512/bundle-manifest.json)

## Bibliographic details
- Authors: C. John McDermott, Paul Cashin
- Published: December 1, 2001
- Series: IMF Working Papers
- DOI: https://doi.org/10.5089/9781451859966.001

---

### Summary of the study
- Univariate studies of the hypothesis of unit roots in real exchange rates have yielded consensus point estimates of the half-life of deviations from purchasing power parity of between three to five years.
- Least squares-based estimates of half-lives are biased downward.
- The paper follows Andrews (1993) and uses median-unbiased estimators of the half-life of deviations from parity as a preferred measure of the persistence of real exchange rate shocks.
- The study uses real effective exchange rate (REER) data for 22 industrial countries in the post-Bretton Woods period.
- Three methods of bias correction are implemented, yielding cross-country averages of half-lives of deviations from parity ranging between 4 to 15 years.
- The REER of several countries display permanent deviations from parity.

### Data and methodology
- Data: real effective exchange rate (REER) data for 22 industrial countries in the post-Bretton Woods period.
- Estimation approach: median-unbiased estimation following Andrews (1993) as an alternative to least squares-based estimators.
- Bias-correction: three methods of bias correction are implemented (methods not enumerated on the landing page).

### Key findings and quantitative results
- Consensus point estimates from prior univariate studies: half-life between 3 to 5 years.
- Bias in least squares: least squares-based estimates of half-lives are biased downward (no numeric bias amount provided on the landing page).
- Bias-corrected results: cross-country averages of half-lives range between 4 to 15 years.
- Some countries’ REERs exhibit permanent deviations from purchasing power parity.

### Subject classification and keywords (as listed)
- Subject: Exchange rate modelling, Exchange rates, Foreign exchange, Purchasing power parity, Real effective exchange rates, Real exchange rates
- Keywords: confidence interval from zero, estimation procedure, Exchange rate modelling, Exchange rates, interval estimate, interval estimation, least squares, least squares estimates of the Augmented Dickey-Fuller, least squares estimator, LS counterpart, LS estimate, Median-unbiased estimation, null hypothesis, Phillips-perron estimate, point estimate, purchasing power parity, Real effective exchange rates, Real exchange rates, shock persistence, time series, unit root model, WP

---

## Content in this bundle

- **An Unbiased Appraisal of Purchasing Power Parity - WP/01/196**
  - [An Unbiased Appraisal of Purchasing Power Parity - WP/01/196 (Markdown version)](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2001/_wp01196.pdf.md){rel="alternate" type="text/markdown"}
  - [An Unbiased Appraisal of Purchasing Power Parity - WP/01/196 (PDF)](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2001/_wp01196.pdf){rel="external" type="application/pdf"}

---

_Source: https://www.imf.org/en/publications/wp/issues/2016/12/30/an-unbiased-appraisal-of-purchasing-power-parity-15512_