Order flows and the exchange rate disconnect puzzle

The aim of this paper is to establish the link between the high frequency dynamics of spot exchange rates and developments in the macroeconomy. To do so, I first present a theoretical model of exchange-rate determination that bridges the gap between existing microstructure and traditional models. The model examines how dispersed microeconomic information known to individual agents outside the foreign exchange market is aggregated and transmitted to dealers via transaction flows (i.e., order flow); and how the information is then embedded in the spot exchange rate. I then report empirical evidence that strongly supports the presence of the link between the macroeconomy, order flow, and high frequency exchange rate returns implied by the model. In fact, my empirical results indicate that between 20 and 30% of the variance in excess currency returns over one- and two-month horizons can be linked back to developments in the macroeconomy. This level of explanatory power is an order of magnitude higher than that found in traditional models — even the newly developed monetary models incorporating central banks reaction functions. Moreover, it provides a straightforward solution to the exchange-rate disconnect puzzle. Namely, the high frequency behavior of spot exchange rates reflects the flow of new information reaching dealers concerning the slowly evolving state of the macroeconomy, rather than the effects of shocks that drive rapidly changing macroeconomic conditions.     

Exchange risk premia and firm characteristics

This paper examines the presence and the determinants of exchange risk premia in stock returns using firm level data from South Korea. We conduct empirical asset pricing tests based on cross-sectional data sorted by firm characteristics such as firm size, liquidity, foreign ownership, and industry. Using alternative model specifications and exchange rate measures, our results support the hypothesis of a significant unconditional exchange risk premium in the Korean stock market at firm and industry levels. More specifically, we find that the exchange risk premium is directly related to firm liquidity and inversely related to firm size and foreign ownership.     

Turnover and return in global stock markets

I study how growth affects liquidity of global stock exchanges and how liquidity determines cross-sectional returns on those stock exchange index portfolios. I measure portfolio liquidity by turnover ratio computed as value of shares traded over the market capitalization. I obtain data from FIBV, an association of global stock exchanges. In a multiple regression model for turnover ratio, I find age, size, type of exchange, competition for order flow, and growth rate to be significant determinants of portfolio liquidity; however, exchange- and time-specific effects are more appropriate for modeling portfolio liquidity. The time effects yield to three distinct regimes, while the exchange-specific effects are surrogates for the legal systems, English common law, and Civil laws of the countries. I estimate the parameters of a multiple regression model in a two-stage GLS framework in which index return is a function of turnover. The GLS method is preferable since a turnover ratio may have a non-stationary, random component. The significant determinants of index return are turnover and volatility, although some of the volatility effect may be a spillover from a January effect. Investors expect higher return from high turnover markets. However, the positive turnover expected return relation is true only in emerging markets; in developed markets expected return is a function of volatility. This result confirms existing empirical evidence that high turnover stock portfolios generate superior returns and further the sources and pricing of risk in emerging and developed markets are different.     

Volatility Spillovers and the Risk-Return Relation Between Stock and Foreign Exchange Markets in Brazil

We examine the evidence of mean and volatility spillovers between stock and foreign exchange markets in Brazil with multivariate GARCH models and nonlinear Granger causality tests. We also use a multivariate GARCH-in-mean model to assess the relationship between risk and return in these markets. The results indicate that the stock market leads the foreign exchange market in price formation and that nonlinear Granger causalities from the exchange market to the stock market do occur. Part of these nonlinear causalities are explained by volatility spillovers. We show that exchange rate volatility affects not only stock market volatility but also stock returns.     

FX Market Returns and Their Relationship to Investor Fear

This note examines the relationship between changes in levels of investor fear (measured by the implied volatility index) and foreign exchange market returns. Our empirical results indicate a negative relationship between daily returns on high‐interest rate (investing) currencies and changes in the implied volatility index, while the association is positive for low‐interest rate (funding) currencies. That is, investing (funding) currencies tends to depreciate (appreciate) when investor fear increases. A sequential breakpoint test identifies a significant change in this relationship in the period following the 2008 collapse of Lehman Brothers, and another in 2012 following the resolution of the European sovereign debt crisis, which suggests that the relationship is linked to financial system liquidity. During the crisis period, currency returns are much more sensitive to changes in investor fear, and this is particularly so for funding currencies that are perceived to present a safe‐haven. The results have important implications for international finance, and those looking to speculate via the carry trade.     

Deep order flow imbalance: Extracting alpha at multiple horizons from the limit order book

We employ deep learning in forecasting high-frequency returns at multiple horizons for 115 stocks traded on Nasdaq using order book information at the most granular level. While raw order book states can be used as input to the forecasting models, we achieve state-of-the-art predictive accuracy by training simpler “off-the-shelf” artificial neural networks on stationary inputs derived from the order book. Specifically, models trained on order flow significantly outperform most models trained directly on order books. Using cross-sectional regressions, we link the forecasting performance of a long short-term memory network to stock characteristics at the market microstructure level, suggesting that “information-rich” stocks can be predicted more accurately. Finally, we demonstrate that the effective horizon of stock specific forecasts is approximately two average price changes.     

China's exports,export tax rebates and exchange rate policy

This paper examines the relationship between China's exports, export tax rebates and exchange rate policy. It offers an explanation for why China's exports continued to rise under RMB real appreciations during the Asian financial crisis. Based on a traditional export demand model, we test our hypothesis that the counteracting effects of China's export tax rebate policy have diminished the effectiveness of real exchange rates in facilitating the resolution of trade imbalances under the current pegged exchange rate regime. We find evidence that RMB real appreciations during the crisis negatively affected China's exports, but the negative effects were mitigated by the positive effects of export tax rebates. We also find evidence of a long‐run relationship between China's exports and the other explanatory variables. The empirical evidence suggests that under the pegged exchange rate regime with limited adjustments, real exchange rate movements alone cannot resolve China's external imbalances. The policy implication of this study is that China needs to redirect its decades‐long export‐oriented development strategy to one that emphasises domestic demand‐oriented development and to replace the current pegged exchange rate regime with a market‐oriented more flexible exchange rate regime.     

Exchange rate jumps and exports: Evidence from China

We apply the autoregressive conditional jump intensity (ARJI ) model to monthly exchange rate returns of China against 81 countries and investigate the impact of exchange rate volatility on exports over the period of 1995–2004. We decompose bilateral exchange rate volatility into continuous and discrete components and find that only the discrete part of exchange rate volatility, that is, the exchange rate jumps, has a significantly negative effect on exports, which to some extent reconciles the old yet unsettled debate in previous literature on the role of exchange rate volatility in international trade. There is also some evidence suggesting that the development of domestic financial market will boost international trade, but it does not help attenuate the negative effect of bilateral exchange rate jump risk on exports.     

我国中央银行货币冲销有效性实证研究

20世纪90年代以来我国国际收支持续双顺差,中国人民银行为了维持人民币名义汇率的稳定和促使宏观经济平稳发展,不断在外汇市场和国内公开市场上进行外汇干预和冲销操作,从而导致了外汇储备持续快速增长、流动性过剩等问题不断出现。通过对我国货币冲销有效性进行实证检验,发现我国外汇冲销效果有限,离完全冲销还有很大差距,分析其中原因,既包括制度、政策因素,也与银行系统内过剩的流动性,存款准备金本身的缺陷有直接关系。为完善我国外汇冲销操作,应增强人民币汇率弹性,促进国债市场发展,完善存款准备金制度。     

Liquidity shocks,commodity financialization,and market comovements

We explore whether and how liquidity factors influence risk transfers between commodity and stock markets using a composite liquidity index and five different types of liquidity measures. We find that liquidity shocks, including both funding liquidity and market liquidity, are positively associated with comovements between commodity and stock markets after 2000, although the relationship is insignificant before 2000. The structural change indicates that financialization creates a role for adverse liquidity shocks to increase cross-market correlations. Further evidence shows that the effect of liquidity on cross-market correlations is state-dependent and intensifies when liquidity conditions deteriorate and asset returns sustain substantial declines. Our findings are not explained by business cycles.     

Returns on ADRs and arbitrage in emerging markets

In this paper we compare the distributions of ADR returns and the returns of the locally traded shares between Chile and Argentina. This comparison is interesting because both countries are emerging economies with a similar free market orientation and the trading hours in both countries virtually coincide with the trading hours in New York. Argentina and Chile differ, however, in two important aspects: During our sample period: (1) The Argentinean market was completely under a fixed-exchange rate system, while Chile maintained a flexible exchange rate regime; and (2) Argentina did not impose any restrictions on foreign investments, while Chile did. We find that the return distributions of the Chilean ADRs are significantly different from the distributions of the returns on the respective underlying Chilean shares. While the mean returns are the same, the return's S.D. are significantly different. In contrast, the hypothesis that the distributions of the returns on the Argentinean ADRs and the returns on their respective underlying shares are the same cannot be rejected. We then use a threshold model to estimate the transaction costs of trading the ADRs and the locally traded shares. We find that the transaction costs that must be added to the returns spread before arbitrage is possible were between 100 and 200 basis points for Chilean ADRs. It was between 66 and 165 basis points for the Argentinean ADRs. The daily return spread reversion caused by arbitrage activities was estimated to be approximately 30% for Chilean ADRs and 40% for Argentinean ADRs. Finally, we cannot reject the hypothesis that low liquidity was a major factor in the cost difference between the two countries.     

Exchange rate management in emerging markets: Intervention via an electronic limit order book

This paper describes and analyzes the implementation of a crawling exchange rate band on an electronic trading platform. The placement of limit orders at the central bank's target rate serves as a credible policy statement that may coordinate beliefs of market participants. We find for our sample that intervention increases exchange rate volatility (and spread) for the next minutes but that intervention days show a lower degree of volatility (and spread) than non-intervention days. We also show for intraday data that the price impact of interbank order flow is smaller on intervention days than on non-intervention days. These stabilizing effects, however, rely on the conditions of large currency reserves and the existence of capital controls; an electronic market seems to support this goal.     

Regime Dependent Sensitivity of Country Exchange Traded Funds to Common Risk Factors

If common factors jointly affect country stock markets, it is an indication of global stock market integration. Common factors may affect some markets more/less than other markets, an indication of the degree of global stock market integration/ segmentation. In this paper, we study the integration of global stock markets based on the returns on exchange traded funds (ETFs) for the US, Canada, UK, Germany, France, Italy, Australia and Japan. The relationship between country ETF returns and common risk factors may be time-varying across countries, and that favors a regime switching (RS) factor model for the dynamics of the country ETF returns. A RS factor model for the relationship between country ETF returns and common risk factors is fitted to daily data for the period from May 31, 2000 to March 31, 2014. We use the data to test a hierarchy of hypotheses on country ETF returns: (1) common factor exposure across all country ETFs and all regimes; (2) common factor exposure across some country ETFs and all regimes, and (3) common factor exposure across some country ETFs and some regimes. The RS factor model for ETF returns fits the data well and the common factors have variable effects across countries and over regimes.     

Liquidity and stock returns in emerging equity markets

Using data for 27 emerging equity markets for the period January 1992 through December 1999, we document the behavior of liquidity in emerging markets. We find that stock returns in emerging countries are positively correlated with aggregate market liquidity as measured by turnover ratio, trading value and the turnover–volatility multiple. The results hold in both cross-sectional and time-series analyses, and are quite robust even after we control for world market beta, market capitalization and price-to-book ratio. The positive correlation between stock returns and market liquidity in a time-series analysis is consistent with the findings in developed markets. However, the positive correlation in a cross-sectional analysis appears to be at odds with market microstructure theory that has been empirically supported by studies on developed markets. Our findings regarding the cross-sectional relation between stock returns and liquidity is consistent with the view that emerging equity markets have a lower degree of integration with the global economy.     

论国际金融市场的利率传导机制

当国际金融市场的利率与一国利率存在差距时,国际资本便会流入或流出,随之又会影响其他国家的资本供给和利率。由于利差的经常存在,国际游资数额巨大,利率差的传导渠道不容低估。从利率的流动性效应着手,运用开放经济条件下利率传导的基本原理及利率、汇率和通胀率之间的关联性,分析国际金融市场利率传导机制的表现,得出来自利率汇率平价机制的市场性传导在金融全球化过程中有力地影响着各国的利率水平,并且通过实际利率的变化影响经济运行步伐。     

Capital market efficiency and the relationship between equity returns, interest rates, and monetary aggregates in Australia

This paper examines the efficiency with which the Australian share market incorporates new information relating to interest rates and the monetary aggregates into share prices. It finds a strong relationship between medium term government security yields and equity returns although little relationship could be found between unanticipated changes in the monetary aggregates and share returns. Furthermore, the interest rate relationship involved long lags and suggests inefficiency in stock market pricing in Australia.     

Why is it so difficult to beat the random walk forecast of exchange rates?

Recent empirical evidence suggests that the time series behavior of the real exchange rate is well approximated by a nonlinear, exponential smooth transition autoregressive (ESTAR) model. This nonlinearity helps resolve a number of puzzles concerning the persistence and volatility of real exchange rates. In this paper, we explore whether it may also help resolve the well-known difficulties of exchange rate forecasting. We develop a bootstrap test of the random walk hypothesis of the nominal exchange rate, given ESTAR real exchange rate dynamics. We find strong evidence of predictability at horizons of 2 to 3 years, but not at shorter horizons.     

Technology-investing countries and stock return predictability

For 77 technology-investing countries we test whether their stock market returns are predictable. We find that exchange rate returns and U.S. stock excess returns predict stock market returns for most countries in our sample, while crude oil and inflation predict returns of less than 40% of countries. While in out-of-sample tests the evidence of predictability declines, U.S. returns still beat the constant returns model for three-quarters of countries in our sample. A portfolio of all 77 countries offers a mean-variance investor annualized profits of between 5.7% and 8.0%, and profits are maximized when return forecasts are based on U.S. returns.     

Liquidity,exchange listing,and common stock performance

The liquidity of securities—the relationship between volume of trading and changes in market price—has won increasing recognition as an element of investment strategy in recent years. Relatively high liquidity is deemed to be a desirable characteristic of a stock, especially for the institutional investor, who typically trades in large volume. Thus, firms can generally be expected to seek means of enhancing the liquidity of their shares. One of the supposed means of accomplishing this is by listing one's stock on a national securities exchange. This paper examines the relationship of common stock liquidity to both exchange listing and price behavior during major up and down movements in the market. Our conceptual and empirical analyses indicate that liquidity is linked to price behavior; and we suggest that the view held by at least some corporate officers—that exchange listing increases liquidity—may be erroneous. More specifically, it appears that when the amount of firm capitalization is taken into account, exchange listing does not result in greater stock liquidity.     

The Impact of Liquidity Risk: A Fresh Look*

This paper takes a fresh look at the importance of liquidity risk using a comprehensive liquidity measure, weighted spread, in a Value‐at‐Risk (VaR) framework. The weighted spread measure extracts liquidity costs by order size from the limit order book. Using a unique, representative data set of 160 German stocks over 5.5 years, we show that liquidity risk is an important risk component. Actually, liquidity risk is increasing the total price risk by over 25%, even at 10‐day horizons and for liquid blue chip stocks and especially in larger, yet realistic order sizes beyond €1 million. When correcting for liquidity risk, it is commonly assumed that liquidity risk can be simply added to price risk. Our empirical results show that this is not correct, as the correlation between liquidity and price is non‐perfect and total risk is thus overestimated.