Foreign Exchange Rate: Private Information In Currency Markets
Imperial College Business School; Centre for Economic Policy Research (CEPR)
University of Cyprus – Department of Accounting and Finance
University of Cyprus – Department of Accounting and Finance
January 22, 2016
Abstract:
Using daily foreign exchange returns for the universe of countries with flexible exchange rates, we document local currency depreciations ahead of public sovereign debt downgrade announcements. Consistent with the private information hypothesis, the effect is stronger in lower institutional quality countries and holds after we control for all concurrent public information and for publically available rumours about the forthcoming downgrades. Our findings persist in countries with no CDS markets further reducing the likelihood that public information from changes in CDS spreads drives the results. Finally, the currency depreciations are permanent, providing evidence for a link between fundamentals and currency markets.
Foreign Exchange Rate: Private Information In Currency Markets – Introduction
In classic textbook treatments, asset prices respond to public news about fundamentals: as public news become common knowledge, asset prices move (for instance, Glosten and Milgrom (1985)). Traditional models of foreign exchange rate determination also posit an information environment where all relevant information is publically known and immediately incorporated into prices and empirical evidence supports the idea that high-frequency foreign exchange rate movements are linked to fundamentals that become common knowledge at pre-specified announcement dates (Andersen et. al. (2003, 2007)). Nevertheless, the link between foreign exchange rate changes and fundamental shocks remains controversial since the seminal work by Meese and Rogoff (1983).
We contribute in this debate by uncovering evidence for a link between sovereign risk and daily, nominal foreign exchange (FX) rates through the behavior of bilateral FX rates against the US dollar before, during, and after, unscheduled official sovereign debt rating announcements around the world.1 We find that in lower institutional quality countries, where information leakage is more likely (Michaelides, Milidonis, Nishiotis and Papakyriakou, (2015), henceforth MMNP), the currency depreciates before the official sovereign debt downgrade announcement, the announcement becomes a non-event (Bhattacharya, Daouk, Jorgenson, and Kehr (2000)), and the depreciation is permanent. Using different measures that control for public information, we build the argument that information leakage reveals fundamental information ahead of the official announcement, thereby providing a direct test of private information in FX markets.
Private information in FX markets is not new in the literature. Lyons (2001) emphasizes the deficiencies of the public information approach and analyses the role of a “dispersed information approach” in FX rate determination (resulting from differential interpretation of public news), where market microstructure and “order flow” play a key role. This approach implies the presence of private information that gets incorporated into prices through order flow. Private information could be associated with dispersed information resulting from differential interpretation of public news, or could be unique information acquired by a group of (large) investors or investors that are followed by other market participants. Evans and Lyons (2008) further emphasize that FX order flow causally affects FX rates and interpret order flow as a proxy for private information that gets transmitted incrementally to FX rates. Consistent with these findings, Ito, Lyons and Melvin (1998) use the abolition of a trading restriction during lunch time in the Tokyo market in 1994 to argue for the presence of private information in the FX market. The recent FX scandal involving the 4 pm fix 2 also illustrates that private information and intra-day trading can materially affect FX rate determination.
To measure dispersed/private information in FX rate markets Evans and Lyons (2008) combine order flow with the Reuters Money Market Headline News (news items observed on screens by market participants), or scheduled announcement of macro variables as proxies of public news. We build on this approach using a different proxy for private information that is motivated by a recent empirical paper (MMNP, 2015) that finds evidence for private information leakage ahead of official sovereign debt rating downgrades. MMNP show that private information affects the local stock market ahead of the official announcement and this reaction is larger for lower institutional quality countries. Private information might leak during the consultation process between the credit rating agency (CRA) and local government officials, it might be intentional or inadvertent, and can also be of either legal or illegal nature (Bhattacharya, 2014). Our empirical analysis naturally addresses two main questions. First, do sovereign debt rating changes reveal fundamental information that affects the FX market? Second, does private information related to forthcoming sovereign debt downgrades move FX markets ahead of the public announcement?
See full PDF below.
