### Abstract

‘Modern’ theories of the Phillips curve inadvertently imply that inflation is an

integrated or near integrated process but this implication is strongly rejected

using United States data. However, if we assume that inflation is a stationary

process around a shifting mean (due to changes in monetary policy) then any

estimate of long-run relationships will suffer from a ‘small-sample’ problem

as there are too few inflation ‘regimes’ where the data are stationary. We offer

a ‘4-stage’ solution to this problem and applying this solution to United States

data we estimate a significant negative sloping non-linear long-run Phillips

curve.

integrated or near integrated process but this implication is strongly rejected

using United States data. However, if we assume that inflation is a stationary

process around a shifting mean (due to changes in monetary policy) then any

estimate of long-run relationships will suffer from a ‘small-sample’ problem

as there are too few inflation ‘regimes’ where the data are stationary. We offer

a ‘4-stage’ solution to this problem and applying this solution to United States

data we estimate a significant negative sloping non-linear long-run Phillips

curve.

Original language | English |
---|---|

Place of Publication | Dundee |

Publisher | University of Dundee, Department of Economic Studies |

Number of pages | 31 |

Publication status | Published - 27 Feb 2016 |

### Publication series

Name | Dundee Discussion Papers in Economics |
---|---|

Publisher | University of Dundee, Department of Economic Studies |

No. | 294 |

ISSN (Print) | 1473-236X |

### Fingerprint

### Keywords

- Phillips curve
- Inflation
- Structural breaks
- Non-stationary data
- JEL Classification

### Cite this

Russell, B., & Rambaccussing, D. (2016).

*Breaks and the Statistical Process of Inflation: The Case of the 'Modern' Phillips Curve*. (Dundee Discussion Papers in Economics; No. 294). University of Dundee, Department of Economic Studies .