Recent developments in microstructure approach to exchange rate determination and implications for monetary policy


In explaining the factors determining foreign exchange rate (FX), three main approaches have been widely used: (i) traditional approach developed before 1970s, (ii) the asset market approach emerged in the 1970s and (iii) the development of market microstructure approach initiated since mid-1990s. The first two approaches are commonly referred to as macro-approach, in contrast, the last one indicates the micro-approach to FX. Recent evidences extensively suggests the link between micro and macro approaches to FX determination. The effectiveness of application of these theories depends primarily on the level of development of financial system and practices of the country. The theoretical background, application of these FX fundamentals and policy implications are discussed below and the lesson-learnt is summarized in this non-technical paper.
The ‘traditional approach’ to FX is known as ‘goods market approach’ in which demand for foreign currencies depend on the purchase and sales of goods i.e. exports and imports of the country.  It basically explains that trade balances influence the FX. In particular, trade deficits (i.e. total imports are greater than total exports) tend to depreciate domestic currency, in contrast, trade surpluses tend to appreciate domestic currency.
In the asset market approach, various forms of the model have emerged since the 1970s and emphasized the role of nominal exchange rate as ‘asset price’ of the durable assets. The first well- known model is purchasing power parity (PPP) theory explanation to FX. It explains that the FX equals the ratio of two price levels (relative price) of related countries in the non-arbitrage condition. In comparison, the uncovered interest parity (UIP) theory explains that FX change is based on the relation between expected returns on short-term interest-bearing assets denominated in different currencies. The ‘asset market approach to exchange rate’ refers to models in which the exchange rate is determined by not only present fundamentals and shocks but also discounted sum of expected future fundamentals and unobserved shocks. In other words, macroeconomic fundamentals of standard exchange rate (FX) theory explains that FX is influenced by the macroeconomic fundamentals such as relative money supplies, outputs, inflations, interest rates and expectation of FX.
Under floating rates, the FX models of asset market approach have explained well the change in FX in line with theory in the long tern. However, the empirical research in mid-1990 have shown that the approach fails to explain short term (i.e., within a week to a year) exchange rate change. The evidence shows that FX movement indicates random walk model, indicating that these models fail to forecast changes in FX in the short term.  Some studies also point out that the most critical determinants of exchange rate volatility are not macroeconomics.
With respect to failure of the asset market approach, two main reasons have been explained on the basis of the assumptions of the model. First, the asset market approach assumes that news on FX fundamentals are publicly available. Based on the arrivals of news and market expectations of future fundamental variables, exchange rates immediately react the effect of these fundamental shifts. Second, this approach assumes that market players are homogenous (i.e., same in taking actions on buying or selling of currency and motives on currency trade) in the FX market.
In contrast, the market microstructure approach assumes that (i) private information, in particular, foreign currency order flow is not publicly available, (ii) market players (participants) are heterogeneous (different) in buying and selling activities and motives on currency trade and (iii) institution such as trading process and trading information is available. The market microstructure approach is based primarily on microstructure finance theories. It is generally defined as the process and outcomes of exchanging assets under explicit trading rules. Two variables (i) foreign currency ‘order flow’ and (ii) spread (bid-ask) constitute major powerful tools of microstructure approach.
Order flow is defined as a measure of buying and selling currency pressure. It indicates transaction volume that is signed. Thus it is the signed buyer-initiated and seller-initiated order for a given time frame. In a dealer market such as spot foreign exchange, the dealers absorb this order flow. In a FX auction market, limit orders absorb the flow of market orders. It is worth noting that order flow determines the FX, however, it is not underlying causes of FX movement. Information/news on order flow is considered as underlying cause.
The recent researches on FX studies have applied the market microstructure approach with high frequency data such as daily to weekly data. The order flow explains significant part of change in nominal exchange rates over periods as long as four months. The evidence of these studies have shown that the results can explain movement of FX not only for the study period so-called ‘in-sample forecast’, but also indicate the forecast for beyond study-period i.e., ‘out-of-sample forecast’ ability for less than a month. The findings show that the FX forecasts under microstructure approach produce better estimates than the random-walk model of asset market approach.
It is worth to note a major difference between two approaches on the view on ‘equilibrium condition’ i.e., equilibrium FX. This is also known as practitioners’ view versus the academic view on equilibrium FX. From the practitioners’ view, FX depreciates because “there were more buyers than sellers.” It means that equilibrium FX is based on these foreign currency supply and demand in the FX market. But macroeconomic view argues that this supply does not include foreign currency of some financial institutions so-called risk-averse investors who do not want to take part in currency trade to avoid risk. Thus FX under microstructure approach reflects ‘transition rate” which is not optimal rate because the buyer/seller imbalance is not known. From macroeconomic view, the FX equilibrium rate can be determined by the supply and demand of all participants in competitive market, (i.e., Walras’s concept of equilibrium) suggesting the need of the ‘Walrasian auctioneer’ in the FX market to reflect the FX equilibrium.
In addition, the current equilibrium FX may change if order flow were observable to dealers in real time (i.e., if the market were more transparent). From a policy perspective, it points to the importance of the effects of increasing order-flow transparency.
(To be continued)

Share this post
Hot News
Hot News