Academic journal article Journal of Money, Credit & Banking

Forward Discount Puzzle and Liquidity Effects: Some Evidence from Exchange Rates among the United States, Canada, and Japan

Academic journal article Journal of Money, Credit & Banking

Forward Discount Puzzle and Liquidity Effects: Some Evidence from Exchange Rates among the United States, Canada, and Japan

Article excerpt

ONE OF THE MOST PUZZLING phenomena of foreign ex-Change markets is that forward discount rates (the difference between forward rates and spot rates) cannot serve as an unbiased predictor for expected changes in spot rates. Furthermore, empirical studies frequently document that the currency of a country with higher nominal interest rates tends to appreciate at later times. Such an empirical finding is dramatically contrary to the standard prediction that forward rates and future spot rates will be jointly depreciated under high nominal interest rates. In the literature on international finance, this puzzling phenomenon is called the forward discount anomaly or the forward discount puzzle.

Serious efforts to resolve the forward discount puzzle have been made in past decades. Hodrick (1987), Engel (1996), and others offer systematic surveys on recent developments of theoretical and empirical research in this field. In one direction, researchers introduce risk-averse behavior into the standard rational expectations model. In another direction, they adopt several alternatives to rational expectations models, including peso problems, irrational expectations, and speculative bubbles. Our reading of the literature is that neither direction has yet completely resolved the puzzle. (1)

This paper empirically investigates the extent to which the explicit consideration of monetary markets can help to resolve the forward discount puzzle. In particular, we examine whether introducing liquidity effects (the negative impact of monetary shocks on nominal interest rates) is able to contribute to solving the puzzle. Intuitively, it is easy to understand that the consideration of liquidity effects is potentially useful for the resolution of the puzzle. Suppose that money supply grows unexpectedly. As a consequence of this, current nominal interest rates decrease immediately due to liquidity effects, whereas inflation rates are accelerated in time to come, due to the quantity theory of money.

The first consequence makes current forward rates appreciate by the covered interest parity. In contrast, the second will cause future spot rates to depreciate by the purchasing power parity. Thus, liquidity effects are able to weaken the one-to-one linkage between current forward rates and future spot rates, thereby offering a reasonable explanation for the action of current forward rates against the movement of future spot rates.

Several empirical studies based on vector autoregression models (hereafter, VAR models) indeed find that monetary policy shocks induce the forward discount puzzle phenomenon. For example, Eichenbaum and Evans (1995) show that a contractionary shock to U.S. monetary policy leads to an immediate increase in U.S. interest rates together with a contemporaneous depreciation of forward rates, while it induces a sharp and persistent appreciation of spot rates. That is, a monetary contraction yields both depreciated current forward rates and appreciated future spot rates. Grilli and Roubini (1996) also find such a monetary-policy-induced forward discount puzzle using VAR models.

These findings suggest that liquidity effects have a chance to play an important role in resolving the puzzle; however, they cannot identify any theoretical mechanism of liquidity effects. It is therefore necessary to examine empirical implications based on structural monetary models in order to investigate thoroughly the relationship between monetary shocks and the forward discount puzzle. (2)

The literature on monetary economics has paid serious attention to the monetary model built by Lucas (1990), Fuerst (1992), and others as a theoretical mechanism of liquidity effects. (3) In their models, households adjust cash positions more slowly than firms, and such an asymmetric cash adjustment generates liquidity impacts on nominal interest rates in money markets. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.