Unofficial Exchange Rates in Managed and Market Regimes with Differing Capital Controls: An Analysis Using Bitcoin
Trinity University – Department of Economics
Themes for the next decade: Cannabis, 5G, and EVs
A lot changes in 10 years, and many changes are expected by the time 2030 rolls around. Some key themes have already emerged, and we expect them to continue to impact investing decisions. At the recent Morningstar conference, several panelists joined a discussion about several major themes for the next decade, including cannabis, 5G and Read More
January 12, 2016
Using bitcoin prices — constructed from daily, publicly available data — I generate exchange rates for 18 currencies with different levels of capital controls and exchange rate regimes. I show that bitcoin-based exchange rates replicate the behaviour of official USD-Euro and the unofficial USD-Argentinian Peso exchange rate (the latter of which has a highly managed official exchange rate) suggesting that bitcoin exchange rates provide daily information on market exchange rates. Additionally, bitcoin exchange rates show fluctuations timed in conjunction with domestic events of the currency’s issuing country, evidence of local participation in bitcoin markets. By comparing bitcoin-based exchange rates to official exchange rates, I evaluate the extent of capital flow restrictions and detect exchange rate regimes of the 18 currencies. Results generally correlate well with the Chinn-Ito Index of Financial Openness and the IMF exchange rate regime classifications, although both these sources will mistakenly classify as free floating or financially open countries that undertake short-term or transitory interventions detected when using daily bitcoin data. In contrast to results from previous literature on managed exchange rate regimes that had been limited to single-country data, I find no consistent pattern of Granger-causality from the unofficial exchange rates to the official exchange rate. Additionally, a proportionality restriction between an unofficial and official exchange rates is satisfied across all exchange rate regimes in the absence of capital controls.
Unofficial Exchange Rates in Managed and Market Regimes with Differing Capital Controls: An Analysis Using Bitcoin – Introduction
Most papers examining unofficial exchange rate behaviour have been constrained to use data from the same data source: the World Currency Year Book (WCYB), formerly known as Pick’s Currency Year Book.1 According to Bahmani-Oskoee et al. (2002), the rates in theWorld Currency Yearbook “are provided by the Central Bank and Ministries of Finance who may be reluctant to provide the true data. Indeed, in some countries that refused to provide the data, the compiler had to adhere to foreign correspondents or informed currency dealers.” I instead use prices from the sales of bitcoin in various currencies to construct exchange rates. As these prices can be directly observed, my method requires no reporting agents removing reporting bias.2 Unlike other on-line sources of currency trades, bitcoin has an alternative use as an investment vehicle. Even if a currency is freely floating there is bitcoin trading activity, and therefore a bitcoin price in the currency. This uniquely allows study of the unofficial exchange rates when the official exchange rate is floating, unlike the experience of Huett et al. (2014) who used an on-line currency trading website to study the unofficial Belarusian ruble exchange rate but found trading ceased once the currency was allowed to float.
I will provide a brief introduction to bitcoin and its history in section (2), and discuss data collection and exchange rate construction in section (3). Bitcoin can function both as a method to bypass capital controls and as an alternative to official, potentially manipulated, exchange rates. Both of these uses can be detected. Intuitively, capital restrictions will be identified by a non-stationary ratio of the two exchange rates, while a peg will be identified by the deviation of bitcoin from the official exchange rate over the sample period. Stationarity shall be evaluated by either the Johansen Trace Test or the PSS Bounds Test detailed in section (4.1). Section (4.2) shows that once trends attributable to bitcoin are accounted for the relative Law of One Price (LOP) applies to bitcoin and official exchange rate trends. As such, deviation from relative law of one price (LOP) provides a estimate of the extent of capital restrictions. These are compared to the Chinn-Ito Index in section (4.3), where differences can be explained by short term, transitory deviations that monthly economic data could not detect.
Exchange rate regime is determined using methods detailed in section (5.1). In section (5.2) I show that while bitcoin has no empirical relationship with the highly managed, official exchange rate of Argentina, bitcoin obeys LOP with the known unofficial exchange rate of Argentina, and can therefore be used to infer the behaviour of daily unofficial exchange rates when the data is not known or reliable. In section (5.3), the regime results are compared to their current IMF classification.
Section (6) re-evaluates current results in the literature regarding proportionality and causality between official and unofficial exchange rates, comparing managed and floating exchange rate regimes with different levels of capital controls. I show that proportionality exists regardless of exchange rate regime as long as capital controls are not implemented. Results from single country studies suggesting Granger-causality from unofficial to official exchange rates in managed exchange rate regimes is not supported. Instead, causality between official and unofficial exchange rates is found to follow no consistent pattern across regimes, even after grouping by the extent of capital controls.
What is Bitcoin? A Brief History
Bitcoin is a crypto-currency designed and created by the entity Nakamoto (2008). A crypto-currency is entirely digital, with no central monetary authority or country of origin. Instead, Nakamoto created 21 million bitcoin, which are discovered by solving mathematical algorithms in a process known as mining. Once discovered, a bitcoin can be held, used for retail purchases, or bought and sold on a bitcoin trading website for a variety of currencies.4 Dwyer (2015) explores value on bitcoin markets and explains why the ability to use bitcoin for these various purposes results in a crypto-currency with a positive price.
A website where bitcoin can be bought and sold is known as an “exchange”. Every account on an exchange has a virtual “wallet” in which the users can store both their bitcoin and the currencies accepted by the exchange. Currency in a wallet can be directly deposited into or withdrawn out of a connected bank account, an online payment system like PayPal, or into a wallet associated with a different exchange. Some exchanges also issue debit cards. An example of an exchange buy/sell interface is shown in Figure (1), using the exchange ANXBTC. A user can buy a bitcoin using currency in their wallet—in the figure, US dollars would be used to purchase the bitcoin—and in a matter of seconds, sell the purchased bitcoin for a different currency—selecting EUR from the drop-down menu would sell the bitcoin for Euros—which would be deposited back into the users wallet. This illustrates why the bitcoin price in US dollars and Euros can be used to construct an exchange rate: Bitcoin can function in a manner similar to a vehicle currency for foreign exchange swaps, a role traditionally taken by the USA dollar.5 Unlike traditional vehicle currencies, access to bitcoin (and its associated exchange rate to other foreign currencies) is impossible to restrict and does not require an intermediate party, as long as agents have internet access. Additionally, bitcoin prices in different currencies across various independent, globally established exchanges are publicly and instantaneously available to all agents regardless of the agent’s country of origin, making bitcoin prices and the associated bitcoin exchange rate difficult to manipulate or regulate, as deviations are quickly removed by arbitrage forces.
See full PDF below.