Bachelor Thesis Rick van Delft

  • Published on
    17-Jul-2015

  • View
    44

  • Download
    3

Embed Size (px)

Transcript

  • Tilburg University

    Stock Market and Liquidity

    Has liquidity increased after the start of the financial crisis?

    Bachelor Thesis Finance

    Rick van Delft

    ANR: 921290

    Supervisor:

    Zorka Simon

    17-05-2013

  • 2

    Index

    Introduction ............................................................................................................................................. 3

    What is the definition of Liquidity? ......................................................................................................... 6

    The level of liquidity ........................................................................................................................ 6

    Liquidity Risk .................................................................................................................................... 7

    The characteristics of stock liquidity ................................................................................................... 8

    How can liquidity be measured? ....................................................................................................... 10

    The bid ask-spread ........................................................................................................................ 11

    The ILLIQ-measure ......................................................................................................................... 11

    The turnover ratio ......................................................................................................................... 12

    What is the expected relationship between liquidity and the stock market? .................................. 13

    What is the role of liquidity in a financial crisis? ............................................................................... 14

    Empirical Evidence................................................................................................................................. 16

    Data description ................................................................................................................................ 16

    The Models ........................................................................................................................................ 16

    The variables ..................................................................................................................................... 16

    The Working Method ........................................................................................................................ 18

    Empirical Results ................................................................................................................................... 21

    Conclusion ............................................................................................................................................. 27

    Bibliography ........................................................................................................................................... 29

    Appendix 1: ............................................................................................................................................ 31

    Amihud & Mendelson (1986) Asset Pricing and the Bid-Ask Spread ............................................ 31

    Appendix 2 ............................................................................................................................................. 34

    Datar, Naik, and Radcliffe (1998) Liquidity and Stock Returns: An Alternative Test..................... 34

    Appendix 3: Results Turnover Rate Model ............................................................................................ 36

    Appendix 4: Descriptive Statistics of the Turnover rate Model ............................................................ 39

    Appendix 5: Results Relative Bid-ask spread Model ............................................................................. 40

    Appendix 6: Descriptive Statistics of the Relative Bid-Ask spread model ............................................. 43

    Appendix 7: Results Bid-ask spread Model ........................................................................................... 44

    Appendix 8: Descriptive Statistics of the Bid-ask spread Model ........................................................... 47

    Appendix 9: Descriptive statistics summary of the Bid-ask spread ...................................................... 48

  • 3

    Introduction

    In times of financial instability, investors are more and more interested in the sources which

    can affect their positions. There are several frictions in the market which can affect the

    investors position. Liquidity is one of them. Within liquidity a distinction can be made

    between the level of liquidity (which is asset specific) and the liquidity risk (which is the

    systematic risk of an asset). The level of liquidity is an important aspect of affecting the assets

    returns. Amihud and Mendelson (1986) state that: liquidity, marketability or trading costs

    are among the primary attributes of many investment plans and financial instruments. A

    simple definition of liquidity is the ability to sell or buy stocks at low cost without affecting

    the price. (Pastor and Stambaugh (2003)) In the same paper they also refer to liquidity as a

    source of systematic risk. Acharya and Pedersen (2005) suggest that there are differences in

    the effect of the liquidity level and the liquidity risk. Because of the costs liquidity is seen as

    an important factor and therefore interest in liquidity has increased. Gomber, Schweikert and

    Theissen (2004) argue its importance because: It affects the transaction costs for investors,

    and it is a decisive factor in the competition for order flow among exchanges, and between

    exchanges and proprietary trading systems.

    Many studies have investigated the relationship between liquidity and the stock market.

    Amihud and Mendelson (1986) studied this relationship between the differences of securities

    bid-ask price on their returns. The difference of the bid- and ask-price is called the spread.

    They find that higher yields required on higher-spread stocks give an incentive to firms to

    enlarge the liquidity of their stocks and thereby reducing their opportunity cost of capital. In

    this way, increasing liquidity can make the firm more valuable. In a later study in 2002,

    Amihud proposes there is a risk premium for stocks excess return. The simple clarification for

    this is because a compensation risk is needed when an investor is exposed to some more risk.

    However, he relates the risk premium also to illiquidity. Illiquidity is the opposite of liquidity,

    thus the higher the spread, the more illiquid a stock is. So, the risk premium is not only

    reflecting the higher risk, it is too a premium for stock illiquidity.

    In the past few decades, many researchers have tried to describe the relationship between

    liquidity and the stock market. They disagree strongly about the best measure of liquidity.

    Kyle (1985) describes market liquidity as: a slippery and elusive concept. So, the factors of

    market liquidity are already hard to define. To measure these factors is thus even harder.

    However, many researches proxy liquidity by using the bid-ask spread. Another measure

  • 4

    often used is the turnover ratio. This ratio can be calculated on several ways which makes it

    known as a measure where many adjustments can be made.

    The aim of this study is to understand the relation between stocks and liquidity and the

    changes in liquidity over the last 15 years. I will use this time frame in particular with respect

    to the financial crisis. Amihud and Mendelson (1986) used data for the period 1961-1980.

    Datar, Naik and Radcliffe (1998) used a sample in the period from July 1962 through

    December 1991. To see whether the same relationship holds for a later period, I will use a

    time-span from January 1997 to December 2012. I will use the same measures of liquidity as

    described in the papers by Amihud and Mendelson (1986) and Datar et al (1998). I have

    chosen for these measures because Amihud and Mendelson (1986) were the first researchers

    with an empirical study about the relationship between liquidity and the stock market. They

    used the differences between the bid-ask price, the spread, to measure liquidity. Datar et al

    (1998) have studied the same relationship, though they used an alternative measure, the

    turnover rate to proxy liquidity. The reason for using this paper is because Petersen and

    Fialkowski (1994) concluded that the quoted spread is not a good proxy for the real

    transaction costs for investors. Thereby, Amihud and Mendelson (1986) give a theoretical link

    between the turnover rate and liquidity. So to test this relationship, I am going to use the

    measure provided by Datar et al (1998).

    The time-span I will use is of interest because of the financial crisis. Brunnermeier (2008)

    studied some of the leading key factors which caused the financial crisis. He also includes

    liquidity as one of the main causes of the financial crisis. He argues that in 2006, many

    investors had put their money in illiquid assets. So they were exposed to liquidity risk, as

    already explained by Amihud (2002). Brunnermeier (2008) and Brunnermeier and Pedersen

    (2009) divided the concept of liquidity into market liquidity and funding liquidity. Through a

    mechanism between these two drivers of liquidity, which later will be explained further, a

    relative small shock can cause liquidity to dry up suddenly and carry the potential for a full-

    blown financial crisis. [Brunnermeier (2008)]

  • 5

    The next chapter will summarize the most important papers about how liquidity is related to

    the stock market. To start with, an explanation of liquidity will be giving. Later the theoretical

    relationship between liquidity and the stock market will be illustrated. Further, liquidity in

    times of a financial crisis will be described. In section 3, the research methodology is going to

    be introduced. The result of the tests will be explained in section 4, whereas section 5 will

    summarize the results and conclusions can be made.

  • 6

    What is the definition of Liquidity?

    The level of liquidity

    Liquidity has many interpretations. The simplest way is to define liquidity as trading an asset

    which could be sold immediately without any costs. Kyle (1985) interprets liquidity as an

    elusive concept of some characteristics of transactional markets which you can distinguish in:

    tightness, depth and resiliency. He states that tightness refers to the difference between the

    bid- and the ask-quotes. Normally for stocks, a bid-price is slightly below the equilibrium

    price whereas the ask-price is higher than the equilibrium price. So if the market is tighter, the

    difference (the spread) between the bid- and the ask-price is smaller and thus the market is

    more liquid. Perfect liquidity could be reached when there is zero spread between the quotes.

    This facet is most commonly used to describe liquidity. However, this measure is only useful

    for trading with low volumes. Larger orders are facing price reactions within the trade, only

    tightness as a measure will not be enough. Thereafter, Kyle uses another concept: depth. Kyle

    explains depth as: the size of an order flow innovation required to change prices a given

    amount. Basically, it is the amount of trades which can be made without affecting the quote.

    Engle and Lange (1997, 2001) have explained depth and tightness in the figure (the market

    reaction curve) below:

    The amount of trading in relation to the depth is here graphically explained by the horizontal

    line as well by the rising line. Low trading volumes do not have an effect on the price which

    is explained by the flat line, whereas the price is rising when the volume of trading becomes

    bigger. The other concept of liquidity is resilience. Resiliency considers the speed of return

    to the efficient price after a random deviation. [Engle and Lange (1997)] Based on the figure

    Source: Engle and Lange (1997)

  • 7

    above, it could be seen as the time it will need to go back to the equilibrium level. There is

    one problem; the equilibrium level has to be estimated. The estimation is very difficult

    because of all sorts of new information could become available in the market, so there is no

    clear point of reference of an equilibrium. This could also be a reason why Kyle states that

    market liquidity is a slippery and elusive concept. Therefore, it is hard to measure liquidity.

    Damoradan (2005) describes liquidity differently. As already said, a simple form of defining

    liquidity is as trading an asset which could be sold immediately without any costs. The

    definition given by Damoradan comes closer to this simple view. Whereas the Kyles

    arguments are more related to the trading aspect and market microstructure, Damoradans

    interpretation of liquidity comes closer to the cost of liquidity, namely:

    When you buy a stock, bond, real asset or a business, you sometimes face buyers remorse,

    where you want to reverse your decision and sell what you just bought. The cost of illiquidity

    is the cost of this remorse.

    Liquidity Risk

    The given definition of liquidity has to do with the level of liquidity. Kyles definitions are

    stock specific, namely what or how liquidity is priced at a transaction. Another distinction

    about liquidity can be made with the systematic liquidity risk. The systematic liquidity risk is

    argued by Pastor and Stambaugh (2003). They investigated the market wide liquidity and the

    correlation of it with the pricing of assets. Acharya and Pedersen (2005) found empirical

    evidence to distract the level of liquidity of liquidity risk. They even argue that: liquidity risk

    contributes on average about 1.1% annually to the difference in risk premium between stocks

    with high expected illiquidity and low expected illiquidity. Although, the whole analysis they

    made was complicated though the collinearity they faced. This is because when securities are

    illiquid, they tend to have also high liquidity risk. Pastor and Stambaugh (2003) identified that

    if the market liquidity drops heavily, stocks returns are correlated in a negative way with

    fixed-income returns. This seems valid with the flight-to-quality, which will be explained

    later in the paragraph of the role of liquidity in the financial crisis. Due to the collinearity in

    the study of Acharya and Pedersen (2005), it is difficult to distinguish the impact of the level

    of liquidity as well as the liquidity risk. The collinearity exists because highly illiquid

    securities often have high commonality in liquidity with the market liquidity, have a higher

    sensitivity to market liquidity and to market returns. Thus, the relationship of all these types is

    complicated to recognize. However, there is another type of risk which has to be mentioned

  • 8

    here, the r...