Working Paper: CEPR ID: DP236
Authors: Mark P. Taylor
Abstract: The covered interest parity (CIP) theorem states that the covered interest differential between two similar assets denominated in different currencies should be zero. This paper utilizes high-quality data recorded by the dealers at the Bank of England to test CIP during periods in which news is known to have introduced turbulence into the market, as well as during a relatively calm, control period. The data is high-frequency and allowance is made for the bid-offer spread, brokerage costs and other considerations. The analysis suggests three important conclusions. First, profitable arbitrage opportunities do occasionally occur and sometimes persist during turbulent periods. Second, the degree of efficiency of the markets we study appears to have increased over time. Third, there appears to be a 'maturity effect' whereby the existence, size and persistence of profitable arbitrage opportunities appear to be a positive function of the length of maturity. We propose an explanation of this phenomenon in terms of the existence of credit limits.
Keywords: covered interest parity; arbitrage; market efficiency
JEL Codes: 431
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Market turbulence (E32) | profitable arbitrage opportunities (G19) |
| Length of maturity (C41) | size of arbitrage opportunities (G19) |
| Market efficiency over time (G14) | frequency and size of profitable arbitrage opportunities (F31) |
| Length of maturity (C41) | persistence of profitable arbitrage opportunities (G19) |