Nonlinear adjustment of exchange rate and exchange rate policy : lessons from Singapore

This paper uses a smooth transition autoregressive regression to estimate the adjustment speed of exchange rate in Singapore, a well-managed small open economy in which exchange rate policy plays an important role in facilitating macroeconomic adjustment and controlling inflation. The result of our...

Full description

Saved in:
Bibliographic Details
Main Authors: Yan, Fangli, Yip, Paul Sau Leung
Other Authors: School of Social Sciences
Format: Article
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/146510
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This paper uses a smooth transition autoregressive regression to estimate the adjustment speed of exchange rate in Singapore, a well-managed small open economy in which exchange rate policy plays an important role in facilitating macroeconomic adjustment and controlling inflation. The result of our estimation suggests that the adjustment speed is faster than estimates reported in the literature. In addition, parallel with the long-run neutrality and short-run nonneutrality of monetary policy in large economies, we provide a pioneer discussion and empirical evidence of long-run neutrality and short-run nonneutrality of exchange rate policy in the small open economy. To our knowledge, we are the first group of researchers conducting rigorous study on this important issue for small open economies. Finally, we found evidence of the Harrold–Balassa–Samuelson (HBS) effect, and our results support Lothian and Taylor's argument that failure to include the effect in the estimation would lead to biased estimate in the adjustment speed. Thus, one should be cautious of the significant portion of past empirical studies that did not incorporate the HBS effect in their estimations. Future empirical studies should include the effect in the estimations and tests, or at least test for the absence of the effect before excluding it from the estimations.