جریان سفارش، اسپرد قیمت خرید و فروش و تراکم تجارت در بازار ارز خارجی
|کد مقاله||سال انتشار||تعداد صفحات مقاله انگلیسی||ترجمه فارسی|
|14856||2012||16 صفحه PDF||سفارش دهید|
نسخه انگلیسی مقاله همین الان قابل دانلود است.
هزینه ترجمه مقاله بر اساس تعداد کلمات مقاله انگلیسی محاسبه می شود.
این مقاله تقریباً شامل 14244 کلمه می باشد.
هزینه ترجمه مقاله توسط مترجمان با تجربه، طبق جدول زیر محاسبه می شود:
|شرح||تعرفه ترجمه||زمان تحویل||جمع هزینه|
|ترجمه تخصصی - سرعت عادی||هر کلمه 90 تومان||20 روز بعد از پرداخت||1,281,960 تومان|
|ترجمه تخصصی - سرعت فوری||هر کلمه 180 تومان||10 روز بعد از پرداخت||2,563,920 تومان|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Banking & Finance, Volume 36, Issue 2, February 2012, Pages 597–612
This paper examines the implications of market microstructure for foreign exchange markets. We argue that the usual order flow model needs to be recast in broader terms to incorporate the transaction costs of liquidity and the limitation of price discovery through order flows that involve low trading density currencies. Using a daily data set, we find that order flows are inadequate when it comes to explaining the changes in the low trading density currencies. Alternatively, within the high trading density, both order flows and bid–ask spreads significantly affect the foreign exchange rate returns. Our findings suggest that the order flow model is better at incorporating these microstructure effects except for some currencies with a very high level of trading density.
This paper examines the implications of market microstructure for foreign exchange markets. We argue that the usual order flow model needs to be recast in broader terms to incorporate the transaction costs of liquidity and the limitations of price discovery through order flows that involve low trading density currencies. Despite the notion of liquidity having previously been applied by a large number of authors to explain the behavior of asset prices, there is no consistent conclusion as to what drives the change in the asset price. Some authors suggest that information transmission is important to price discovery in the short run (Evans and Lyons, 2002 and Evans and Lyons, 2008). However, others attribute the change in the asset price to the liquidity (Roll, 1984, Glosten and Milgrom, 1985, Huang and Stoll, 1997 and Chordia et al., 2008).1 On the one hand, empirical investigations of the order flow model question the ability of the information transmission to engage in foreign exchange (FX) rate discovery through excess buying (selling) pressure.2 That is, when orders to purchase (sell) the base currency exceed orders to sell (purchase) it, the corresponding FX rate rises (falls).3 Several studies support the view that the order flow is to a significant degree successful in explaining the change in the FX rate in the short term, and possibly even in the medium term. On the other hand, liquidity refers to the matching of buyers and sellers. It is inter-temporal in nature and is not necessarily linked to price discovery. As a simple example of this distinction, suppose that all buyers of an asset arrive on Thursday and all sellers on Friday. In this case, no trade will take place on either Thursday or Friday in the absence of counterparty. To solve this problem, it is possible to introduce the liquidity supplier who will sell to the buyers on Thursday and buy from the sellers on Friday. To provide this liquidity, a spread emerges between the ask and bid prices to compensate the middleman. Recently, through the mediation process described in O’Hara (2003), a dispute between the two parties was settled. O’Hara (2003) clearly points out that these two concepts are related, but they are not the same for price discovery. Hence, when one wants to know what drives the change in the asset price, both information transmission and liquidity should be included as explanatory variables. As each function can influence asset prices, we first discuss how the order flow conveys the information to the FX rate markets, and then consider how liquidity enters into the determinant of FX rate. Order flow is a major determinant of FX rates because it conveys the information through the excess buying (selling) pressure. Evans and Lyons (2002), for example, in their seminal paper report that a regression of daily exchange rate returns on daily order flow results in an R-square in excess of 60%. Following Evans and Lyons, 2002, Evans and Lyons, 2008 and Bacchetta and van Wincoop, 2006 employ the macroeconomic information collection system to discuss how the indirect effect of the information transmission via the order flow affects the volatility of the FX rate. However, any such issue is possibly misguided when the order flow fails to explain the low trading density currency due to the limitation of price discovery. The limitation of price discovery indicates that the price impact of the order flow should be positively correlated with the trade-size clustering. That is, the order flow of the active (high trading density) currency tends to be obviously larger than that of the infrequently traded (low trading density) currency. Evans (2011) points out that, without a great deal of trading activity, dealers will not immediately adjust their spot rate quotes via the order flow to accommodate the new information on fundamentals and the risk of providing liquidity to the market. As a result, the order flow is not suitable for investigating the behavior of FX rates when currency pairs trade with low liquidity. The limitation of price discovery, as the measure of the market quality, is very important to the investigation of the order flow. In particular, some academic research articles argue that the explanatory power of the order flow should be positively associated with it. For example, Moulton (2005) and Gwilym and Meng (2010) highlight the relationship between the price impact of the order flow and trade-size clustering in order to explain the variations in the common liquidity measures of foreign exchange transactions. Their consistent conclusions show that the price impact of the order flow is greater when customers (traders) care more about trading precise quantities, while the number of trades and the total volumes are not significantly changed. These challenges indicate that the order flow provides neither a model nor empirical estimates of the “liquidity based” and “clustering of transactions” price impact factor. Chan and Fong (2000) therefore assert that the volatility–volume relationship underlying stock returns could in principle be driven by either one or both components of the number of trades and the size of trades.4 Moreover, Kaul and Sapp (2009) shed more light on the issue regarding the effect of the trading activity and dealer concentration on foreign exchange market quality. They clarify that the trading activity is an important determinant of market quality and dealer concentration.5 According to the above evidence, liquidity measures not only affect the volatility–volume relationship but also the information transmission ability of the order flow. We therefore suggest that the liquidity is another wedge of the microstructure that investors should look for in FX markets. Liquidity reflects the ability to buy or sell significant quantities of an asset both quickly and anonymously. To maintain liquidity, several exchange systems use liquidity suppliers, i.e., individuals who stand ready to sell or buy whenever the customer intends to buy or sell. As suppliers of liquidity, they have monopoly rights to buy at the lower bid price and sell at a higher ask price. This privilege is the primary source of compensation for offering liquidity, namely, the bid–ask spread. 6 It is more important to notice that transaction data occurring in financial markets are often contaminated by market microstructure effects, such as bid–ask spreads, liquidity ratios, turnover, and asymmetric information. Market microstructure effects indicate that changes in transaction prices over very small time intervals are mainly composed of noise and carry little information about the underlying return volatility (Hasbrouck and Seppi, 2001 and O’Hara, 2003). Amihud (2002) also point out that the bid–ask spread partly represents the illiquidity premium on the expected stock return due to the small trading size. With the same meaning, we expect that the liquidity suppliers should require more bid–ask spreads on low trading density currencies to compensate for their illiquid risks. According to these arguments, the presence of market microstructure effects is necessarily accounted for in the order flow model. This study’s first objective is to compute Stoll’s realized bid–ask spread as the measure of the liquidity (Stoll, 1989). Such a spread is markedly different from the usual spread because it is easy to detect the pattern of price reversal and price continuation in a transaction. Alternatively, the usual spread is driven solely by the relative gap between the ask price and the bid price. We estimate the coefficients on the regression of the daily exchange rate return on the bid–ask spread, and order flow for each currency pair. We find that the order flow is a dominant explanatory variable for active currencies, while the bid–ask spread is a potential candidate in explaining the daily returns on infrequently traded currencies. For the purpose of the comparison, our second objective is to provide a Generalized Method of Moments (GMMs) to understand the link between the order flow and the information dispersion, which is captured by the estimated coefficient on the variance of the order flow. Moreover, we also examine whether the variance in price changes can be attributable to the deviation in the bid–ask spread when trading density is at a high or low level.7 We show that the contribution of information dispersion is an important determinant of the variance of exchange rate returns for active currencies. However, the empirical results for infrequently traded currencies yield the opposite conclusion because most of these currencies are insensitive to the clustering of transactions. The remainder of this article is organized as follows. In Section 2, we introduce two common liquidity measures in economic and financial studies. In Section 3, we describe the data and the empirical procedures of the extended order flow model including the daily realized bid–ask spread. In Section 4, we present our results, and in Section 5, we conclude. The Appendix reports the robustness checks of the GMM model.
نتیجه گیری انگلیسی
This paper examines the implications of market microstructure for foreign exchange markets. We argue that the usual order flow model needs to be recast in broader terms to incorporate the transaction costs of liquidity and the limitation of price discovery through the order flow such as low trading density currencies. We measure the buying (selling) pressures by using the order flows and measure the liquidity by using the realized bid–ask spreads. Using a daily data set with a distinct trading density, we find that the order flows have significant influences on the FX rate returns for high trading density currencies rather than for low trading density currencies. Alternatively, the bid–ask spreads also play nontrivial roles in the determination of the FX rates, especially for the low trading density currencies. This finding is evidenced by the positive association between the non-adverse selection components of the bid–ask spreads and the trading density. We therefore conclude that the risk associated with information-based trading is lower for high trading density currencies than for low trading density currencies. We also implement the lagged model to investigate whether this strong positive relationship is appropriate to the lagged order flow and the lagged realized bid–ask spread. Consistent with Sager and Taylor (2008) and Cerrato et al. (2011), our results show that the estimated coefficients and adjusted R-squared value of the lagged variables have now declined drastically. The coefficient on the lagged order flow is statistically insignificant in all currency pairs regardless of the high and low trading density. These empirical results seem to suggest instead that the micro-model approach as in Evans and Lyons (2002) might be questionable when we are not allowed to connect the lagged order flow and bid–ask spread to the contemporaneous FX rate return. Using the GMM model, we also find that the information dispersion of the FX rate return is greater when trading density is higher. Overall, our work sheds light on the importance of the liquidity in the usual order flow model. Moreover, the information transmission of the order flow tends to be fruitless when the trading density is low.