Quantifying the effects of new derivative introduction on exchange volatility, efficiency and liquidity

This thesis investigates the effects of the introduction of new financial derivative products on exchange volatility, efficiency and liquidity. The derivatives under primary investigation are Exchange Traded Funds (ETFs) and Contracts for Difference (CFDs). These products offer a cheap, tax-efficient and speedy method for increasing or decreasing market exposure to price changes in the related primary asset. By facilitating faster and shorter-term trading, these products may increase market liquidity and/or increase market volatility for the related primary asset. The thesis builds a cross-country database of new-derivative-markets opening dates, and investigates the key features of prices and returns for related primary assets before and after the opening of these derivative markets. The database covers 16 countries in the CFD investigation, 21 commodity markets in the ETF investigation, and related data as available (daily closing prices, trading volumes, bid-ask quotes) in each of them. The key price and return features investigated include bid-ask spreads, trading volumes (both of derivatives and related primary assets), and daily return autocorrelation, variance, skewness and kurtosis. This thesis also considers a separate, but related, research problem. It extends and empirically applies a liquidity-indicator model for the Eurozone created by the Bank of England (BOE) and developed further by the European Central Bank (ECB) by including commodity liquidity, and uses this extended model to investigate shifting investor behaviour based on changing market dynamics. Similar to the investigation of the CDF and EFT markets, this investigation is concerned with market stability and liquidity in a changed environment (in this case, the key change is the introduction of the euro currency). Chapter one contains an introduction to the main hypotheses regarding the effects of the introduction of new derivatives on securities markets, and the empirical methods used to test these hypotheses. This chapter also describes the two investment products which are the main focus, CFDs and ETFs, and their particular potential impacts on market-specific characteristics such as volatility, efficiency and liquidity. Chapter two empirically investigates the impact of CFDs on market liquidity and volatility. CFDs have existed for less than twenty years and the CFD market grew rapidly prior to the recent international financial crisis. This chapter empirically examines the roles that CFDs have played, either as an accelerant for mispricing in international equity markets away from fundamental values, or as a source of increased market efficiency through the addition of new liquidity. This chapter uses GARCH and EGARCH models to test for the impact of CFDs on the return volatility and autocorrelation of the underlying security. In the case of Australia, the analysis is applied to individual securities. In the other 15 countries investigated in this chapter, the analysis is applied at the level of the market index. The chapter also investigates whether the stylised characteristics of CFDs are more or less pronounced in low liquidity exchanges. This chapter finds that CFDs appear to have influenced asset-specific variance and return autocorrelation. Some tentative explanations for these findings are offered. The presence of bid and ask-price ‘overhangs’ associated with CFD trading cannot be rejected and may be associated with observed volatility reductions in some jurisdictions. Following the analysis based on CFDs in chapter two, ETFs are the primary focus of chapter three. ETFs have existed since the late 1980s, but were first traded on commodity markets in the early 2000s. Their inception has been linked by some market analysts with the large growth in commodity market volatility seen in recent years. This chapter directly tests this link. The chapter also investigates whether the stylised characteristics of ETFs are more or less pronounced in larger commodity markets than in smaller markets. The results indicate that larger ETFs in terms of their assets under management at their dates of inception, are associated with higher volatility. Smaller commodity markets are found to have increased efficiency after the introduction of ETFs, indicating that there are some benefits from new ETF investment in markets below $4 to $5 billion in size, but the associated caveat is that of increased volatility, indicative of potential pitfalls in the ETF portfolio rebalancing process. It appears that ETFs have made commodity markets more efficient through a new influx of trading counterparties, but they appear to be associated with a cost. The need for regulation of investment size and market ownership limits therefore cannot be rejected. Chapters two and three look at two particular new instruments and their effects on liquidity and volatility. Another major innovation in market structure was the advent of the euro currency in January 1999. The power and presence of a financially-combined Europe attracted new international investment, therefore influencing liquidity. The combination of this influx of investors and new products (including CFDs and ETFs) can potentially have wide market impacts. Understanding the structural changes of liquidity in Europe in recent years is important for macroprudential risk assessment, as sudden changes in conditions may be indicative of current stress and a signal of future stress. Chapter four presents a Europeanspecific liquidity measure used by several central banks, and provides some new modifications to this measure. The measure is constructed by combining several facets of liquidity and depth measurement across several asset markets. It attempts to incorporate aspects such as market tightness, depth and resiliency. The flows and the direction of causality can also be inferred using vector autoregression, Granger causality techniques and impulse response functions. The measure uses a combination of liquidity determinants including the bid-ask spread, the return to volume ratio and numerous measures of liquidity premia. In the chapter, the modified liquidity measure is applied empirically to European-area data