2017 International Conference on Mathematics, Modelling and Simulation Technologies and Applications (MMSTA 2017) ISBN:

Transcription

1 2017 International Conference on Mathematics, Modelling and Simulation Technologies and Applications (MMSTA 2017) ISBN: The Impacts that Margin Trading has on the Liquidity of Underlying Stocks: An Empirical Research Bo ZHANG 1, Pan LIU 1 and Allen YANG 2 1 Xi'an University of Technology, Xi'an China, University of Toronto, Toronto Canada, M5S 2E8 Keywords: Liquidity of underlying stocks, Margin trading, Expectations inertia. Abstract. Under the background of increasing margin trading scale in Chinese stock market, a research that its influence on the liquidity of underlying stocks is made in this thesis. Through theoretical analysis, we assume that there exists different expectations inertia model. The influencing process of margin trading on the liquidity of underlying stocks is also deduced to be subdivided as transaction channel, volatility channel and participation channel. By using panel data regression model, the empirical analysis is made to test the actual effect of margin trading system on the liquidity of underlying stocks in Chinese A share market. Empirical results show the transaction channel and volatility channel hypothesis are fully verified in the empirical study of financing transaction system on the liquidity of underlying stocks, while participation channel hypothesis is partly verified. The further analysis based on the view of institution designing and participant structure of the market is also made and the corresponding recommendations are also suggested in this paper. Introduction Henry and Mckenzie (2006) studied short-term trading data in the Hong Kong stock market from 1994 to 2001, and found that short selling exacerbated market reaction to new information, resulting in a "polarization" of stock returns. Chang (2007) used event study to analyze the price fluctuation of the stock in the short list, and found that the price fluctuation tends to rise when the stock is short. Leung and Liu (2015) use Granger causality test find that margin trading will increase the volatility of A-share market, but there are significant differences in the ability of financing transactions and margin trading because of different institutional design. Arturobris (2003) found that, compared to markets where margins are not allowed the frequency distribution of negative returns in the margin market is likely to be reduced by cross-market comparison. Therefore, he concludes that margin trading can play a role in stabilizing stock price volatility. Hong and Stain (2003) have investigated whether margin trading is preventing stock price declining by introducing a "heterogeneous agent model" and found that prohibiting or restricting margin trading would induce excessive market declines. Xu Xiaoguang et al. (2013) measured the impact of margin trading on the volatility of the Shanghai and Shenzhen markets through the ARCH model. The results show that after the implementation of margin trading the volatility of Shanghai and Shenzhen market has declined and conclude that margin trading can stabilize the market. Dai Qing et al. (2014) simulated the non-linear mechanism of micro-buying and selling behavior of investors through the artificial stock market model (ASM) of multi-agent based on Swarm platform from the point of view of computational experimental finance. The results show that, under the current trading system, margin trading can restrain the fluctuation of stock price to some extent. Xiao Hao and Kong Aiguo (2014) use the quarterly panel data which selected form November 2011 to January 2013 under the subject of two margin financing to build DID model for testing, and find that margin trading can reduces the non-information efficiency factors and the volatility of the stock price. Based on the GJR-GARCH model, Sohn and Seo(2015) found that decreasing the short selling limit can reduces the market asymmetric volatility. Sun Hao (2011) have a research on the influence of margin on the turnover of 567

2 A-share market in 120 trading days from March to September in 2010 which take the financing buy-in amount, the financing balance changes amount, the margin changes amount, the total amount of margin change as the margin variable agent. It is concluded that financing transactions can affect the size of market transactions, but the short selling transactions cannot. Based on the experimental financial approach, Fellner and Theissen (2014) found that the existence of margin trading restrictions can not only weaken the impact of disagreement on stock prices, but also expand the bid-ask spread and reduce the size of the transaction. By reviewing the related literatures, a prevalent conclusion can be suggested that domestic and foreign scholars generally tend to think that a margin on the underlying stocks have a certain influence on the liquidity, however, have a divergence on the direction of influence. Existing research is mainly focused on the effect of margin trading on share price volatility and then how it affects the liquidity. It is also meant that the margin (the input side) and the underlying stock liquidity (output side)which carrying a lot of intermediate variables and intermediate reactions are within a black box that give rise to the mechanism of the margin on the liquidity of underlying stock. We can conclude that the existing results about the Mechanism of the Effect of Margin on Stock Liquidity still lack of systematic research on the basis of theoretical analysis. In general, it is difficult to smooth the divergence between the different conclusions of the study, because the impact of margin on the underlying stock liquidity mechanism of black box is still not fully uncovered. Theoretical Analysis The change of stock price is affected by many factors such as economy, society and system, but it is directly caused by the trading behavior of market traders. Trader's trading behavior is based on individual information acquisition, knowledge processing, experience, and formed on the basis of individual cognition, through the group interaction process to amend or strengthen their cognitive judgments, and then formed the relatively persistent and stable psychological expected. This kind of psychological expectation includes the expected value of the stock intrinsic value and the external value. The expected performance of the intrinsic value of the stock is a kind of value judgment, that is, a reasonable range of the intrinsic value of the stock, and underestimated compared with the market price, Overestimation, or correct valuation. But the external value is more expected to appear as a price trend judged that the stock price trend of the psychological determination of the future. In some certain expectations, investors tend to adopt the appropriate trading strategies and trading systems in the existing trading activities. On the one hand, the trading behavior changes the market participation degree of the investor group through the interaction, on the other hand, it directly acts on the stock supply and demand, and induces the stock price fluctuation and transaction scale change, and finally has a comprehensive effect on the stock liquidity. While the changes in stock prices become part of the new part of the impact of the trader after a part of the cognitive, expected, trading behavior thus forming a cycle of cycles. In the meantime, if the one-way strengthening is expected to be maintained, there will be unilateral strengthening of transaction behavior, market participation, price fluctuation and change of transaction size, which will lead to the stock liquidity of the subject to increase or decrease. On the contrary, if attenuation or reversal is expected, it will trigger the reverse transformation of trading behavior, and then induce the price trend reverse, format the underlying stock liquidity reverse in the same mechanism. Hypothesis 1: Securities lending and borrowing through the "trading channels" affect the underlying stock liquidity. We believe that margin trading can expand the supply and demand of stock trading based on the credit basis, and then transform the potential transaction demand induced by heterogeneous value judgments based on different information acquisition and information interpretation to actual transaction demand. It is manifested that whether the prices rise or fall, traders with optimistic value judgments can participate in stock transactions and increase stock demand through financing transactions and traders with pessimistic value judgment can form the stock supply through the short 568

3 selling transaction. With the increase of capital and stock entering into the transaction process, the depth and width of the underlying stock transaction will tend to rise, which will reduce the price shock induced by the stock exchange and then raise the underlying stock liquidity level. Hypothesis 2: margin trading through the "wave channels" affect the underlying stock liquidity. Margin trading based on different expected behavior patterns has an impact on underlying stock price that fluctuates and then affect the liquidity level. The impact and basis of margin changes induced by the degree of stock are determined based on premise of the transaction behavior of the expectation of traders. It is expected that the inertial characteristics of its own are presented as the process of formation, diffusion, strengthening and mutation of non-linearity. The expected inertia process is distinguished from the direction of the expected momentum phase, while the prediction of the mutation is in the expected reversal phase. Accompany with margin trading,short purchase, under the expectation that the asset value is undervalued, leads to the rise of stocks,this validate and strengthen the previous expectation and then exhibit expected momentum effect which lead to further short purchase which again lead to price and volatility rise and put pressure to the liquidity of underlying stocks, this is confirmed by Xu Hongwei and Chen Xin (2012).On the other hand, if stocks rises continuously by a large margin, which convert to the opposite direction of underestimation of price, that is the price overestimation expectation, in this way expected reversal effect can lead to the decrease of short purchase and then the price fall which overhaul the previous rise of stock price and ease the volatility, which give backstop to the liquidity of the underlying stocks. Similarly, expected momentum effect leading to short selling with overestimation expectation will intensify stock volatility. While expected reverse effect caused by continuous price falling will ease short selling and volatility of stock price so as to backstop liquidity. Hypothesis 3: margin through the "participation channels" affect the underlying stock liquidity Regularly disclosed margin data to a certain degree can convey the expected information by other traders to the underlying stock values that include price judgments and price trends to the market traders. Because of the incompleteness of the traders themselves and the incomplete validity of the market, there is a possibility of interaction among traders, which may induce changes in investor sentiment that are more time-varying than expected. (2012). This has been confirmed in studies based on computational experimental methods. The positive information conveyed in the scale of financing transaction or the shrinkage of margin size tends to form investors' optimism. Optimistic sentiment tends to net buy or hold equity assets, which lead to higher investor participation in the underlying stocks and affect the trading posture and drives the stock up. On the contrary, the reduction of financing scale or the increase in the size of the stock market will induce the pessimism which tends to lead to the decline of investor participation and falling of shares. That is the way called "participation channels" that affect the underlying stock price and liquidity by influencing the investor sentiment and investor participation. Data and Variables In order to reveal the impact of margin on the stock liquidity in different market conditions and considering July 2015 from margin trading margin may trigger data distortion, we select from January 2014 to June 2015 a total data of 18 months, 385 trading days for the study period. To reveal the possible difference of the impact of margin on the stock liquidity of different attributes, according to the daily average market capitalization during the study period, the sample stocks were divided into three groups: low-flow market value sample (less than 10 billion RMB), medium-market capitalization sample (10 billion-200 billion) and high-flow market value sample (over 20 billion). According to theoretical analysis and modeling needs, we select financing transactions, margin trading, transaction size, investor sentiment, stock price volatility as an explanatory variable, select the liquidity as an explanatory variable. The relevant variables are listed below (see Table 1). 569

4 Table 1. The list of variables. variable Proxy variable Calculation formula Attributes Financing transactions Balance of daily financing (MP) Positive Transaction size Balance of Japanese securities (SS) Positive Securities trading Daily Turnover (TV) DCF it ( Pit NAVU it ) / NAVU it Positive Investor sentiment Closed-end fund discount rate (DCF) Reverse Stock price Exponentially weighted moving 2 2 volatility average volatility (VOL) ˆ t t 1 1 rt 1 Positive LIQ ln p ln p / TM / TMV Reverse fluidity Liquidity Index (LIQ) t t 1 Empirical Analysis A panel data regression analysis Low-flow market value of the underlying stock is made below. First, a random effect regression model is established. Second, using the Hausman test to determine whether the model should be selected from fixed-effect model or random effects model. The W statistic of the panel data model was (P = 0.05), therefore, we reject the null hypothesis that the individual effect in the random effect model is not related to the explanatory variable under the significance level of 5%, further we consider that there is less possibility of individual influence and structural change among the cross-sectional members, So the establishment of fixed-effect constant coefficient model regression test, test results in Table. 2. Table 2. Low-flow market value of the sample stock comprehensive test table. dmp SS VOL 0.469***( ) ( ) TV ***( ) *** ) DCF -23.2***( ) ( ) The similar Regression Analysis are also made, the results are shown from Table3 to Table 6. Table 3. In the circulation of the market value of the sample stock comprehensive test. dmp SS dvol TV DCF LIQ *** *** *** -6.83*** ** Table 4. In the flow of market capitalization of the sample sub-channel ticket check. dmp SS dvol ***( ) ( ) TV ***( ) *( ) DCF -23.2***( ) ( ) Table 5. High - flow market value of the sample stock comprehensive test table. dmp SS dvol TV DCF LIQ *** -5.84** *** -1.01*** Table 6. High - flow market capitalization sample stock channel test. dmp SS dvol *( ) ***( ) TV ***( ) ( ) DCF -1.26***( ) ( ) Based on the above three sets of sample tests, it can be found that the "transaction channel" hypothesis has been confirmed by the fact that the financing transactions in all three groups of samples through the positive impact of the transaction size to influence the underlying stock liquidity. The "Volatility channel" hypothesis has also been confirmed by the fact that financing transactions in all three groups of samples also through a significant positive impact on stock price volatility have 570

5 influence on the underlying stock liquidity. Although the financing transaction in the three groups of samples has significant positive effect on investor sentiment, which indicates that the financing transaction has positive information transfer function. It only affects the sample stock liquidity significantly in the sample of market capitalization, and we can concluded that the "participation channel" assumption in the financing transaction is only partly confirmed. The margin trading has a significant positive impact on underlying stock liquidity in the three groups of samples, indicating that it has the same function of raising the underlying stock liquidity as the financing transaction. Nevertheless the sub-channel research is showing a significant difference. In the low and mid-market sample, the short-term trading affects the underlying stock liquidity only by positively influencing the size of the transaction, which suggests that the "channel of trade" hypothesis only exists in the sample of the low and mid-market, and the "volatility channel" hypothesis, which has been confirmed in high-flow market capitalization samples, does not hold in the medium and low-flow market samples. In addition, the "participation channel" assumption of margin trading is not true in all three groups of samples. Comparing the margin of securities with the underlying stock liquidity, we can found that, as the financing of the credit transaction, the margin trading has a positive effect on the liquidity of the three sample stocks through "trading channels". The "wave channel" of financing deals is smooth, but it can put negative pressure on the underlying stock liquidity. Because the positive effect of the "trading channel" represented by the size of the transaction to maintain a positive impact on the liquidity of the underlying stock. Margin trading tends to increase trading volume while reducing stock price volatility, thus help to boost the underlying stock liquidity by means of "trading channel" and "wave channel". The "participation channel" hypothesis in margin trading is rarely confirmed (except for the financing transactions in the sample stocks of the circulating market capitalization), in which the financing transaction itself can significantly affect the investor sentiment, while the margin trading cannot. It shows that the financing transaction is compared to the securities trading for investors with a strong function to transfer information. The empirical results show that margin trading can effectively improve the underlying stock liquidity level, which will promote the quality of China's securities market. At the same time, there are asymmetric effects of margin on the underlying stock liquidity through which the influence of financing transactions on the underlying stock liquidity is more stable and diversified, while the influence of short selling on the underlying stock liquidity is single and devoid of stability. The reason can be explained from the following aspects. First of all, there are flaws in the design of margin financing system. The performance of the existing financing channels are more diversified and smooth, While the short selling securities, due to the lake of smoothness of securities financing system and heterogeneity of underlying securities which lead to a limited amounts and types of securities, are restricted form the expansion of the size of the margin, the resulting phenomenon is that the practice of financing long-term scale is much higher than the size. It is, to a certain extent, inhibited the role played by the securities lending mechanism, and formatting of margin on the underlying stock liquidity asymmetric impact. This can also explain why the financing transaction has strong information contents, and the securities exchanges with the value of information that is subject to system constraints cannot truly reflect the market demand for potential margin, and then the value information distorted. Secondly, there are differences that intrinsically linked to the trader structure and stock attributes of China's stock market between margin groups in different market capitalization. In the two sample groups with low and high market capitalization, investor sentiment has no significant effect on liquidity. The reason is that the low-flow market value sample stocks are mostly small-cap stocks, and their institutional degree of investors is relatively high, so less affected by investor sentiment. In the two sample groups with low market capitalization and high market capitalization, investor sentiment has no significant effect on liquidity. 571

6 Conclusions This paper builds the conceptual model on the basis of combing domestic and foreign literature, and puts forward the hypotheses that the margin trading is based on three channels: transaction channel, fluctuation channel and participation channel. We select the 323 trading days from January 2014 to March 2015 as the researching period, take 214 stocks in the A-share market as the researching samples, and then divide them into three groups according to the market capitalization size in a bid to carry out empirical tests through the establishment of panel data model. The results show that margin has a significant positive effect on the stock liquidity, which plays a role in improving liquidity, but the impact mechanism is different. It manifests that the "trading channel" and "fluctuating channel" hypothesis of the financing transaction in the sub-channel test are confirmed, and the "participation channel" hypothesis is confirmed in some samples. The diversification and stability of the impact path are shown as a whole, which manifest that the "transaction channel" and the "fluctuation channel" hypothesis in the margin trading are partially confirmed, while the "participation channel" assumption has not been confirmed. Acknowledgements This study has been funded by Shaanxi Natural Science Foundation (2017JM7015). References [1] Anchada Charoenrook, Hazem Daouk. A study of market-wide short-selling restrictions [J]. Working paper, Vanderbilt University, 2005 [2] Arturobris, William N. Goetzmann, Ning Zhu. Efficiency and the bear: short sales and markets around the world [C].Yale International Center for Finance, [3] Chang E., Chen J., Yu Y. Short-sales constraints and price discovery: evidence from the Hong Kong market [J]. Journal of Finance, 2007(62), [4] Conrad, J. The price of effect of option introduction [J]. Journal of Finance, 1994(44), pp [5] Dai Qin, Xie Fei, Yan Guang-le. Study on the Influence of China's Margin System on Stock Market Fluctuation Based on Swarm Platform [J]. Shanghai Economic Research, 2014(9): [6] Dong Chen-yu, Zhao Pei, Liu Wei-qi. Research on the Function of Margin in Improving the Quality of Stock Market [J]. Management Modernization, 2014(6): [7] Fang Libing, Xiao Binqing. Effect of Margin Unbalance on the Pricing Efficiency of Underlying Stocks [J]. Journal of Contemporary Economic Science, 2015(2): [8] Fellner G., Theissen E. Short Sale Constraints, Divergence of Opinion and Asset Values: Evidence from the Laboratory [J]. Journal of Economic Behavior & Organization, 2014, 101(1), pp [9] Figlewski, Stephen, Gwendolyn P. Webb. Options, short sales and market completeness [J]. Journal of Finance, 1993(48), pp [10] Grube, R. Corwin, O. Maurice Joy, and John S. Some empirical evidence on stock returns and security credit regulation in the OTC equity market [J]. Journal of banking & finance, 1987(11), pp [11] Henry, O.T.T., M. Mckenzie. The impact of short selling on the price-volume relationship: Evidence from Hong Kong [J]. Journal of Business, 2006(79), pp

7 [12] Hong, Harrison, Jeremy C. Stein. Differences of opinion, short-sales constraints and market crashes [J]. Review of Financial Studies, 2003(16), pp [13] Kong Xiangyu, Bi Xiuchun, Zhang Shuguang. Effect of margin on liquidity: an empirical study based on China's stock market transaction data [J]. Enterprise Economics 2014(6): [14] Kraus, Alan, Amir Rubin. The effect of short sale constraint removal on volatility in the presence of heterogeneous beliefs [J]. International Review of Finance, 2003(4), pp [15] Liang Xingyun, Liu Weimin. Study on margin trading and stock market volatility in China [J]. Friends of Accounting, 2015(18): [16] Lin Xiangyou. Effects of Margin Trading on ETF Fund Market Quality-Based on Double Difference Model [J]. Investment Research, 2014(6): [17] Ma Jing, Wang Zhi-yuan. Effect of margin on liquidity, risk and volatility of underlying stock-empirical research based on double difference (DID) model [J]. Finance of Hainan, 2014(6): [18] Robert H. Battalio, Paul H. Schultz. Options and Bubble [C]. EFA Maastricht Meeting Paper, 2004, pp [19] Sohn P., Seo J.Y. Investor heterogeneity and asymmetric volatility under short-sale constraints: Evidence from Korean fund market [J]. Estudios De Economia, 2015, 42(1), pp [20] Sun Hao. Empirical Analysis of the Effect of Securities Margin on Stock Market in China [J]. China Economic and Trade, 2011(6): [21] Sun Zanzan. Study on the Impact of Margin on the Performance of China's Securities Market [D]. Donghua University, [22] Woolridge J.R., Dickinson A. Short selling and common stock prices [J]. Financial Analysts Journal, 1994, 50(1):