“Meese-Rogoff Redux: Financially-weighted Exchange Rate Forecasting,” June 2007.
This paper re-examines the Meese-Rogoff puzzle using the
financially-weighted exchange rate and information on the external imbalance
and financial account. In particular, I focus on whether new information on the
financial account would help predict exchange rates at short and long horizons
and what would be the source of noise to disconnect exchange rates and economic
fundamentals. It shows that information on the external imbalance cannot beat
the random walk at horizon shorter than one quarter, but it plays a role in
predicting exchange rate changes at longer horizons. The predictive power is
mainly driven by the trade imbalance rather than the asset imbalance. These
findings are consistent with limited asset valuation adjustment in
“Measuring Korean International Portfolio Positions, Rates of Return, and Financially-weighted Exchange Rates,” February 2006.
This project works to construct a data set for Korean international portfolio positions, rates of return, and financially weighted exchange rates at quarterly frequency since 1980. One contribution of this work is to construct portfolio positions at market value by interpolating end of year positions and including valuation changes arising from asset price and exchange rate movements. Another contribution is to distinguish four types of assets (direct investment, equity, debt and other investment) and their currency compositions. This distinction is particularly important for the analysis of international adjustment, because different assets pay different returns and are exposed to different exchange rate risks.