Explaining Apparent deviations from Covered Interest Parity: Evidence from Mexico
DOI:
https://doi.org/10.21919/remef.v18i1.830Keywords:
Covered interest parity, vector autoregression models, vector error-correction models, forward premium, exchange rate, funding liquidity.Abstract
This paper tests and quantifies the effects of reduced funding liquidity conditions on the covered interest parity (CIP) relating the U.S. Dollar-Mexican Peso market. To this end, a vector error-correction model is estimated. Results suggest, first, that apparent deviations from the CIP disappear when measures of funding liquidity for market participants are considered. Second, the exchange rate forward premium and the U.S. interest rate adjust towards the CIP cointegrating relationship. Finally, a structural analysis shows that deviations from CIP are mostly determined by shocks on the funding liquidity in the U.S. while funding liquidity conditions in Europe also have a non-negligible role. From the policy perspective, the paper underlies the relevance of funding liquidity measures when assessing whether the foreign exchange market works efficiently. As ever, there are some caveats in the analysis to consider. First, funding liquidity measures may shift from non- to stationary regimes. Second, market participants may not able to fund their liquidity at reference rates. The financial series present considerable ARCH-like behaviour, this may be a source of information to explore in further work.

