Stock Market Reactions To Bank Loan Announcements

In this thesis we will investigate whether stock price reactions of small firms listed on the New York Stock Exchange are larger than stock price reactions of large NYSE listed firms after a bank loan announcement is made.
The role of banks as financial intermediaries in today's economy is significant. Banks play a vital role in connecting institutions that would like to borrow money, the borrowers, with institutions that are willing to lend, the depositors. As an intermediary, a bank has several roles to fulfill of which the most important one has just been mentioned. An intermediary is delegated the task of costly monitoring of loan contracts written with firms who borrow from it (Diamond, 1984). Schumpeter (1939) stated that the banker must not only know what the transaction is which he is asked to finance and how it is likely to turn out, but he must also know the customer, his business and his private habits. By doing this, bankers know much more about a company then any other possible financial player. This reduces the information asymmetry between banks and potential borrowers.
Another role banks play is 'creating' money and regulate the domestic and international payment systems. Well-working payment systems are essential to an efficiently performing economy.

The reason why firms want to borrow money from banks is to fulfill payment obligations or to invest in projects and actions that will help in establishing the goals a firm has set. A bank loan to a firm has several conditions, from which the most important one is that the bank monitors the firm. The most important reason for a bank to monitor the firm is to calculate the risk of the loan and to reduce moral hazards at the firm (Diamond, 1991). This means banks want to minimalize information asymmetry as much as possible to ensure that the loan is favorable for the bank as well and to make sure the company is able to repay the loan. Banks solve these asymmetric information problems by producing and analyzing information and by setting loan contract terms, such as the interest rate charged or the collateral required, to improve borrower incentives (Berger, 1995). Empirical studies such as James (1987) and Lummer and McConnel (1989) found that, in the 1980's, when there was a bank-borrower relationship, the value of a company was raised.
Relationship lending helps reduce asymmetric information between lenders and borrowers. Banks are better able to evaluate the risk of default for current borrowers, and these current borrowers can benefit from more favorable borrowing terms and better credit availability over time (Kysucky and Norden, 2013). Normally when one speaks of relationship banking, three options must be present. (1) A bank gets information which is not publicly available, (2) receiving information happens over a longer period of time, through extensive contact with the borrower, and (3) the information stays confident (Berger, 1999). Especially the latter, is a perfect example of the advantage banks, as financial intermediaries, have over other potential investors. We can expect that stocks of smaller firms react stronger to bank loan announcements then large, public firms. Information about large firms namely is better publicly available than information about small firms. So when a bank issues a loan to a smaller company of which there is little information present for 'outside' investors, this will result in trust, which will result in a higher stock price.
Bank loans have always been important to firms. Fama (1985) and Diamond (1991) state that bank-lending activities provide firms a certain quality and this signals creditworthiness to outside investors. Lummer and McConnel (1989) state that banks play an important, and perhaps, unique role as transmitters of information in capital markets. Banks put lending decisions on the basis of information that is not available to other capital-market participants. Fama (1985) states that banks have better and more insight into the quality of the company that public investors do not have. The decisions that these banks make provide signals about borrowers' creditworthiness. Bank loans also result, due to monitoring effects, in more transparency, which means it is easier to counter adverse-selection. Diamond (1984) also states that an interesting implication of the delegated monitoring model is that intermediary assets will be illiquid. This is because the intermediary is delegated the task of observing information about each loan which no one else but the entrepreneur/borrower observes.
In this thesis we will look at the effect a bank loan itself will have on a company's stock price. As stated above borrowers are monitored by banks, the lender. This monitoring reduces potential moral hazard problems firms may have (Diamond, 1991). The past decades there has been a lot of research to investigate the reaction of stock prices after bank loan announcements. In these researches there was found empirical 'evidence' of mostly positive stock price changes as a result of bank loan announcements. Mikkelson and Partch (1986), James (1987), and Slovin et al. (1988) report positive effects on client firm value in response to announcements of credit agreements, bank loans, and letters of credit. Stock markets play an important role in (most) economies, so a positive effect on stock prices and thereby the entire stock market is important for (almost) every economy and therefore everybody that is directly or indirectly related to economies around the world (that will eventually be about everybody around the world). Levine and Zervos (1996) underlines this, by stating that stock market development is positively associated with economic growth.
Slovin et al. (1992) find that there is a different effect on stock price changes between large firms and small firms. Slovin et al. conclude that for small firms both renewals and initiations of loan agreements generate significantly positive share price effects. A renewal of a bank loan means that an existing loan will get replaced with a new agreement. An initiation of a loan agreement is a completely new bank loan agreement with a firm. Lummer and McConnel (1989) states that when bank loan agreements are divided in new agreements and renewals, only positive loan renewals have a significant excess return. While only the new agreements have almost no impact on stock prices. For large firms Slovin et al. (1992) find little evidence that bank credit agreements convey information to the capital market. These results are consistent with arguments of Fama and Diamond that primarily small, less prestigious firms receive benefits from screening and monitoring services associated with bank loans.

In our thesis we will further investigate the effect of bank loan announcements on small NYSE listed companies compared to the effect of bank loan announcements on large NYSE listed companies. The hypothesis therefore will be: 'Stock price reactions of small NYSE listed firms are larger than stock price reactions of large NYSE listed firms after a bank loan announcement is made'.
This paper expands on the work of Slovin et al. in 1992 by dividing the firms into small and large firms and look at the differences in the CAR between these two groups. We will research the more recent years, namely 2000-2012, to examine whether if there are significant differences between earlier studies and the results we find.
From a manager's point of view little abnormal return can be favorable due to the fact that managers mostly are judged on the company's value, which is measured by the value of one stock multiplied by the total outstanding stocks. This beside the fact that many managers own personal stocks of the company, to avoid agency problems, hence it then would be of their own personal interest to gain a high abnormal return.
We will compare the stock price reactions for small and large firms by calculating the two-day average excess return of the stock price after a bank loan announcement is made. By using this method we can observe the difference between the predicted (normal) return and the actual return and, whether this is indeed a larger positive effect in the case of small firms.
The remainder of this paper is organized as follows. Section 2 contains the literature review, section 3 the data & methodology. In section 4 you will find the multivariate regression analysis. Section 5 contains the meta-analysis and the last section; section 6 contains the conclusion and lessons learned.

II. Related Literature

Approximately the past 30 years there has been a lot of research on the topic of bank loan announcements and it's effect on stock/security prices.
Banks have a comparative cost advantage in making and monitoring repeated short-term inside loans (Black, 1975). In 1984, Diamond explains that this financial information, monitored by financial intermediaries (like banks), is information that is only available to the intermediaries and borrowing firms (so not to 'outside' parties like e.g. shareholders). This conclusion raises the question whether actions and decisions of banks concerning their clients are based on this inside information. Fama (1985) underlines this and adds the statement that actions like loan issues by banks signal information about the creditworthiness of a borrower. We presume banks preferably lend money to companies with low risk of default. This assumption, information and statements together imply that when a loan is issued, this will only occur if the intermediary like for instance a bank is convinced that the borrower is creditworthy and has a low risk of default. So when a loan is issued, this logically is a positive and thus desirable sign for every shareholder of a borrowing company, for it is less likely that their stocks are becoming less valuable. It's a positive signal that can be brought down to the value of one shareholder's stock(s) of the company. So subsequently, it would be logical reasoning that when a loan is issued (and announced), this positive sign of a borrowing firm's position results in a higher demand of this firm's stocks and thus a higher price of these stocks.
The first real event study where an effect of a (private) credit agreement on stock prices has been found, finds an average positive abnormal effect of 0,89% on stock prices in a sample of 88 NYSE listed firms between the years 1972-1982, and in a two-day window after a bank loan announcement has occurred (Mikkelson and Partch, 1986). In the same year, Freeman (1986) comes with conclusions that 'information supporting common stock prices systematically differs between large and small firms' and that 'the magnitude of abnormal return associated with good or bad news from a common class of signals is inversely related to firm size'. This is the first time firm sizes become a subject in this context. Since bank loans are 'good news', we expect share prices of small firms to react more positively than with large firms. One year later, Collins et al. (1987) state that especially at small firms, stock prices contain information about the expected future performance of these firms. In the case of raised stock prices after bank loan announcements, we suggest that the information that is captured in this raised price (that suggests positive expected performance), is information that is directly related to and originated in the inside information of the lending banks about the related company. More on firm sizes is discussed later on.
Shortly after these studies, more studies appear researching whether there is an effect of bank loan announcements on stock prices. James (1987) finds an average positive abnormal effect of 1,93% on stock prices in a sample of 80 NYSE listed firms between the years 1974-1983, and in a two-day window after a bank loan announcement has occurred. He also finds negative effects of -0,91% and -0,11% in stock price after an announcement of private placement or public straight debt respectively has occurred. James explains that there are some factors that provide the abnormal returns. The factors James mentions are maturity, borrower's default risk, borrower size, purpose of borrowing and credit rating. However, these factors do not solely provide the abnormal returns found. This again strongly suggests that there is something special about bank loans that is not equaled by other lenders, and that therefore especially lender type is an important determinant for the reaction of stock prices after an announcement of some kind of loan has occurred. This suggestion can be combined with explanations of Billet et al. (1995) and Fama (1985). Billet et al. (1995) explain that 'if lenders obtain private information in the process of underwriting loans, their lending decisions would then convey valuable information about a borrower's true risk'. Second explanation comes from Fama (1985), who states that lenders may have different monitoring abilities, which enhance a borrower's value by assuring that appropriate investment and spending decisions are implemented. These two explanations in combination with the results of James (1987) (plus earlier studies as discussed before) thus explain that when a bank decides a loan is issued, that this decision is based on inside information and certain monitoring skills that other lenders do not equal, and that this is a signal to the stock markets, who do not have this information, that the company is expected to perform well in the future (otherwise the bank would not issue the loan at all).
More studies again find average positive abnormal effects in two-day windows. Lummer and McConnell (1989) find a CAR of 0,61% in a sample of 728 NYSE listed firms between the years 1976-1986, and Slovin et al. (1992) find a CAR of 1,30% in a sample of 273 NYSE listed firms between the years 1980-1989. Lummer and McConnell (1989) are the first to introduce the concept that loan type is a relevant factor to take into account. In their study, they find new loans to induce no abnormal returns at all, where for revisions the mean abnormal return is 1,24%. They furthermore confirm '['] it is the action of the bank, rather than the borrower's decision about the use of debt, that signals information'. Lummer and McConnell conclude their study with two important statements that underline almost every precedent study: '['] decisions made by banks as a result of a continuing lending relationship with a corporate borrower serve as influential signals of firm value. Thus, the results indicate that banks are important and credible transmitters of firm-specific information to the capital market'. Remarkable about this final conclusion is the fact that advantages of 'inside information' as discussed before, are not mentioned. In line with this is, it is claimed that banks do not possess any 'competitive advantage over other lenders in making credit decisions at the outset of a loan'. This contradicts the image created before, where one could say that different monitoring skills create competitive advantages.
In 1990, Carter and Manaster introduce another dimension that influences the insight in the processes related to this topic. Underwriter's (like a bank) reputation is introduced, and there is stated, '['] low dispersion firms will attempt to reveal their low risk characteristics to the market. They do this by selecting prestigious underwriters. Prestigious underwriters only market ['] low dispersion firms'. This reasoning suggests that more prestigious banks are related to less riskier loans, and thus would it be logical that a bank loan announcement of a higher rated bank would be understood as a stronger positive signal about a borrower's future.
As indicated before, firm size is an interesting factor to take into account concerning this topic. In this study, we are mainly interested in to what extend the firm size influences the effect on stock prices after a bank loan announcement has occurred. Hereby we are mainly interested in whether this is still the case, if this has been the case previously anyway. The first real event study to examine this has been conducted by Slovin et al. in 1992. Slovin et al. find an average positive abnormal effect of 1,92% for a sample of 156 small NYSE and NASDAQ listed firms and 0,48% for a sample of 117 large NYSE and NASDAQ listed firms (all from 1980-1986). This difference is remarkable, and interpreted as 'evidence that banks provide valuable monitoring and certification services primarily for small capitalization firms'. Slovin et al. (1992) claim that these results are consistent with Fama arguing that 'Information on an ongoing deposit history also has special value when the borrower is a small organization (or individual) that does not find it economical to generate the range of publicly available information needed to finance with outside debt or equity', and 'I suggest that for individuals and for some organizations, especially small organizations, that do not have outside equity, the contracting costs for inside loans like bank loans are lower than for outside debt' (Fama, 1985). These two statements suggest that small firms are more likely to be financed for the largest part by inside debt (like bank loans) than large firms would be (who thus would be more likely to be financed for the largest part by outside debt (like public stocks)). Altogether, the statements of Slovin et al. (1992), Fama (1985), Freeman (1986) and Collins et al. (1987) combined imply that greater positive abnormal effects on bank loan announcements with small firms might be explained by a suggestion that the more an organization is financed by bank loans (in comparison with outside debt), the larger the effect on bank loan announcements will be, and since there's suggested that small firms are financed by bank loans in a greater way, small firms will have greater abnormal effects on bank loan announcements. In 1992, Becketti and Morris tried to come to an answer on the question whether bank loans are still special. This question on itself indicates that by then, the assumption that bank loans where special was taken for granted. They conclude their study by stating that bank loans in fact are still special, especially for small and medium-sized firms. However, they also mention that 'bank loans are less special than they used to be', and that 'good substitutes for bank loans have become more available'. This is the beginning of a turnaround in vision about bank loans. More on this turnaround vision is discussed later on.
In 1993, Best and Zhang conclude in their study that banks only examine inside information of firms they lend to when these firms are expected not to perform well in the future. They mention that only in these situations, loan decisions are based upon inside information. They also state that '['] banks gain access to more 'inside' information about the firms with whom they deal. While this may be the case, the mere access to 'inside' information does not imply its utilization'. This implies that when a loan announcement is made in the case of a firm that is expected to perform well in the future, that this decision is almost completely based upon public available information. If this would be true, then abnormal returns on stocks from firms that are predicted to perform well in the future would not be based on some 'special service' or role of banks. Question then is why the abnormal returns in this case occur, we suggest that for investors the new loan might function a wake-up call that their shares are still going strong. Best and Zhang (1993) also find results that indicate that when a loan announcement is made in the case of a negative earnings forecast, abnormal return on stocks are much larger than when the forecasts are positive (in comparison with the negative forecast is this return negligible). There are also differences in returns when different kinds of loans are issued, like exactly is the case in Lummer and McConnell's study (1989): Best and Zhang's study gives a strong indication that favorable revisions and especially mixed revisions induce larger abnormal returns than new loans.
For now, we have identified a few disturbing factors that seem to influence the abnormal returns after bank loan announcements. These factors are: firm size, borrower rating, bank rating, forecasts on performance of borrowing firms, purpose of borrowing, type of loan and maturity. Furthermore, is seems that the 'special' role of banks fades a bit.
Billet et al. (1995) conclude that 'the existing literature clearly indicates that announced bank loans generate positive returns for the borrower'. We agree on this, but not in every situation, like we have pointed out before. Billet et al. themselves also find an average positive abnormal effect of 0,628% on stock prices in a sample of 540 NASDAQ listed firms between the years 1980-1989, and in a one-day window after a bank loan announcement has occurred. However, when they separate the cases into firms that received loans from AAA rated lenders and BAA or lower rated lenders, Billet et al. find an average positive abnormal effect of 0,636% and an average negative abnormal effect of -0,571% respectively. This is consistent with the already mentioned conclusions of Carter and Manaster (1990), and confirms that not only lender type is very important to incorporate in every analysis concerning bank loan announcements but also lender's credit ratings.
In 2002, Hadlock and James come with some interesting statements 'In particular, the results suggest that for firms with public debt outstanding, it takes a high degree of undervaluation in the public securities markets for the firm to cross the threshold where the information benefits of bank debt finance outweigh the relative contracting costs'. Hadlock and James (2002) hereby assume that only under evaluated firms will go for bank loans instead of other lenders. This is an important assumption to take into account. Furthermore, Hadlock and James suggest that returns are affected by credit conditions, which is in line with what Fama (1985) stated before.
Billet et al. (2006) wrote another remarkable article about bank loans, where they investigate whether bank loans are special. In this article, there is mentioned that private loans and straight debt issuance are associated with negative subsequent performance of the borrowing company over a period of 3-5 years. This 'challenges the view that private lending agreements provide unique benefits for the borrowing corporations. ['] In fact, this paper provides a strong image that reductions in information asymmetry are unlikely to explain the positive short-run returns associated with loan announcements'. Furthermore, 'Borrowers have lower operating income and capital expenditures than their peers in the year of and following the event, and this continues for two or three years after the loan has been put into place'. These results are 'consistent with the poor stock returns' of these companies. Billet et al. are pretty sure however that banks are aware of this poor performance, and that they are unlikely to share in this poor performance. All these results indicate that the market is systematically wrong about the perceived direction of the event's effect on firm value going forward. Thus, conclude Billet et al., 'bank loans are not special'. This conclusion is radically different from the perspectives of earlier studies treated before. We did mention that we found Becketti and Morris (1992) talking about a 'turnaround vision'. In the article of Billet et al. in 2006, this turnaround is clearly assumed completed. In the same year, Fields et al. (2006) presented a study called 'Do bank loan relationships still matter'?. The first sentence in their conclusion strongly supports the conclusions of Billet et al. in 2006, they state: 'We find strong evidence to suggest that bank loan announcement returns have become insignificant in the years following studies by James (1987) and Lummer and McConnell (1989)'. This conclusion has been derived from Fields et al. their test results. They found that abnormal returns where positive during the 1980's (0,76% for the first half and 0,44% for the second half) and 1990's (0,80% for the first half and 0,26% for the second half), but become 'insignificant' by the period of 2000-03 (0,13%). Their study was conducted with examination of average abnormal effects on stock prices of 1111 U.S. firms found with Nexis/Lexis, in a two-day event-windows in groups of 5 years starting from 1970-1990 after bank loan announcements. Apart from the first half of the 1980's, there clearly is a descending trend to be observed from around the mid 90's. The small effect in de last period of 2000-03 may be an almost negligible effect, but it is an effect.
Fields et al. also found that 'renewal loan announcement returns are more likely to be positive for smaller companies', which is in line with earlier studies of Fama (1985), Slovin (1992) and Becketti & Morris (1992), and found different effects on loan type which are in line with the earlier studies of Lummer & McConnell (1989) and Carter & Manaster (1990).
In 2009, Lee and Sharpe conducted a study wherein they used a new measure that proxies for monitoring. This measure is able to predict future loan quality and determines loan pricing (Coleman et al., 2006). Lee and Sharpe (2009) find some confirming results in their study, and draw a new view on the whole topic. They find a 'positive relationship between the bank's loan screening and monitoring and the borrower's loan announcement standardized cumulative abnormal return'. This is an interesting new factor to take into account but also important, as Lee and Sharpe point out: 'screening and monitoring proxy has greater explanatory power than several ex post proxies such as bank size and credit rating that have been used in prior studies'. The abnormal return found in their study is 0,92% in a sample of 201 DJNS listed firms between the years 1995-1999. They furthermore again confirm that a bank's screening and monitoring of a borrower adds value to the borrower. However, they also state that abnormal returns induced by a relatively high level of screening and monitoring is economically relatively small. Finally, they also find that the loans used to refinance existing debt add value to the borrowing firm, which is related to Fama (1985) describing purpose of borrowing as an influencing factor.
In 2010, Ross finds an average positive abnormal effect of 1,03% on stock prices of 1064 U.S. firms found with Factiva between the years 2000-2003, and in a two-day window after a bank loan announcement has occurred. This is strongly in contrast with Fields' results for the same period. Ross again finds, like Lummer and McConnell (1989), Carter and Manaster (1990) and Fields et al. (2006), that different types of bank loans such as revisions or new loans are important to take into account. Finally, Ross states that lenders credit rating and status are a very important determinant in this topic, he states that: 'dominant banks are 'special' precisely because they are better at banks' canonical roles of screening and monitoring'. This logically leads to the assumption that the loans researched by Lee and Sharpe (2009) merely have been issued by dominant banks.
Gonzales (2011) has conducted a recent study that brings some very interesting conclusions. At first, she confirms that 'banks can gain access to private information that is not available to other credit claimants', which is in line with several other studies as extensively discussed before. Furthermore, she mentions the fact that almost every public firm has a bank loan, and states that only 22% of all bank loans are reported. Also, she finds that loans with a longer maturity, bigger loans and more risky loans are more likely to be reported. Gonzales however also states that 'since the cost of reporting has decreased with technological developments, there is an increase in the frequency of reporting during the 1996 through 2004 study period that makes the information content of the average loan decrease overtime'. This may be related to results Fields et al (2006) and Billett et al. (2006). She concludes by stating that 'Still, and consistent with the view that bank loans are more informative about firm potential when they are reported in the press, reported borrowers improve their operating performance with respect to non-reported ones over the three years following the activation of the loan'. So in the end, it seems that bank loans announcements still signal positive information to the market.
The literature as extensively discussed in this literature review provides a wide view of the development of and the forces related to this turbulent topic. All relevant related literature that is available has been discussed. Numerous strong forces all have their influence on stock market reactions to bank loan announcements, as explained before. In this study, we are primarily interested in the relation of differences of abnormal returns and firm size nowadays. The existing literature clearly indicates that small firms have had greater average positive abnormal effects on stock prices than large firms roughly for the period between 1985-1995. The question however rises whether this is still the case for NYSE listed firms nowadays. Recent (contradicting) studies make us question whether bank loan announcements still have an effect on stock prices, and if so, whether there is an economically relevant difference between small and large firms (if there's a relevant effect at all).

In figure 1 on the next page, there can be observed that all event studies we have discussed before are shown together. We will discuss some theories about possible explanations for the differences between the studies.

#: Study: N:
1 Mikkelson et al. (1986) 360
2 James (1987) 80
3 Lummer and McConnel (1989) 728
4 Slovin et al (1992) 676
5 Billet et al. (1995) 540
6 Aintablian and Roberts (2000) 122
7 Fiels et al. (2006) 1111
8 Ross (2010) 1064

In figure 1, the horizontal axis represents the average abnormal effect on stock prices. The vertical axis has no value but attaches a number to each study in the figure. The blue dots represent the average abnormal effect on stock prices per study, and the lines extending to both sides of them together represent the confidence interval.

The combined confidence interval can be observed at Y=9, has a mean value of 0,842% and goes from 0,565% to 1,119%. This is a reasonable precise confidence interval. Also can be seen that all studies have overlapping intervals, which means that they are in some way consistent with each other. These studies are more or less comparable with each other, and the small differences in outcome may be declared by small differences in time frame, source, sample, methods etc.
The studies are arranged by year of publication and thereby also for the greatest part by period examined in each study. Based on figure 1, we cannot say that CARs are decreasing over time. However, Field et al. find a decreasing effect in their study.
The prediction interval is the interval that provides us the range where effects in next studies in the same setting are expected to come out. This interval naturally is an interval only based on the studies included in figure 1. The fact that this prediction interval only includes positive values is of managerial relevance. Based on figure 1, it means that managers can expect a positive reaction on their company's stock prices after bank loan announcements at all time, or at least not a negative reaction. This would mean that bank loans are at least in favor when compared to other financing methods. Further examination on the development over time of returns after other types of financing is needed to say which one is the best. Nonetheless, bank loans are positive, based on figure 1. Since decisions between different financing types are very different, have a high impact and therefore are very relevant, these conclusions are of a very high managerial relevance.
Also, the results as showed in figure 1 are of economical relevance. All CARs are positive and on average just less than 1%. This may seem a small effect but in fact this is a large effect, for most CARs found are based upon relatively very large firms. Every single stock of these large firms has a price increase of just less than 1% on average, which means a huge amount of money for all stocks together. This thus happens to relatively large firms who dominate and pretty much determine the economical situation. Every change in their performance or value is economically relevant, especially an effect that is close to a change of 1%.
The only study that really is pretty different is the study of James (1987). It is very remarkable that the study of James (1987) has such a wide confidence interval, where all other studies have more precise confidence intervals. Also, the effect size found by James is very large in comparison with other studies. This is especially remarkable when James his results are compared with the results of Lummer and McConnel (1989) and Mikkelson et al. (1986). The latter two, namely, are results of studies conducted for almost exactly the same time period, and are much smaller than James finds in his study. We suggest this difference can be declared by many different factors like for example different stock markets, information sources, different calculations, possible combinations of different loans, etc.
Another study that is slightly different in comparison to the rest of the studies is the one of Aintablian and Roberts (2000), but we expect the higher result to be declared by the fact that this study examines only Canadian firms. We thus expect it is very important to only examine firms of one single country.
What cannot be seen in Figure 1 is the fact that Fields et al. find a rather small effect size for the period of 2000-2003 (0,13%). This effect size forms a large discrepancy compared with the effect size found for the period of 2000-2003 by Ross (2010) (1,03%). We discovered that Fields et al. use NexisLexis in their study, where other studies use a variety of other sources. We suggest that NexisLexis significantly provides different data in comparison with most of the other possible sources. Some other studies used databases like Factiva, DealScan, Compustat, etc. We will use the ThompsonOne (T1) database. We expect these different databases to differ sometimes dramatically. It would be better for all studies to use one standard database. The exact differences may be an interesting topic for studies examining the differences between information sources regarding financial data, but we will not examine this thoroughly.
The discrepancy between different studies motivates us to re-examine the past thirteen years.
So in order to create a clear image of the effect of bank loan announcements nowadays, we will re-examine the effect of bank loan announcements with the size of firms taken into account in this thesis for the years 2000-2013.

III. Data & Methodology

We selected our samples of bank loan announcements from the ThompsonOne (T1) database. We looked for public data for the period of 01/01/2000 ' 03/01/2014, for firms located in the United States, and in any industry. We examine this time frame because this time frame shows different results among different studies (for example, discrepancies can be found between Fields et al (2006) and Ross (2010)). We examine our own results in order to try to provide more detailed information for this period and come with answers on how this is possible. Furthermore, we are interested in results that are more recent. The other search criteria include: company names, loan initiation status, loan package amounts, loan announcement dates, years to final maturity, Moody's and Poor's credit ratings and total assets.
The ThompsonOne database provided us from 19,544 bank loan announcements over the chosen time period. We adjusted the loan announcements dataset by eliminating all dates that were incorrect, which left us with at 18,654 announcements. After this, we corrected the dataset by eliminating the companies for which no total assets were provided at all, which left us at 13,873 announcements, incorrect CUSIP8 codes, which left us at 12,604 announcements, and double loan announcements that left us at 9,086 announcements. After merging the dataset with data from Eventus, where we also set Eventus for only selecting NYSE/AMEX listed firms, 7,820 announcements were still present. After the merger we also found out that the Cumulative Abnormal Return (CAR) for 2013 was actually not provided at all, so we excluded those announcements as well, that left us 7,132 announcements. At last we had to filter the to eliminate all the missing values of the sample. Finally after the filtering we finally came to a sample of 402 bank loan announcement for the period 2000-2013 that are left to examine.
To calculate the cumulative abnormal return we use the market model event study methodology. We are using an event study because we are investigating the causal relationship between the independent variables (bank loan announcements) and the dependent variable, stock price. This research strategy observes the values of the independent variables and suggests the causality because the value of the dependent variable is measured immediately after the event. The downside of an event study is that it is impossible to avoid the influence of a third (independent) variable on the outcome of the dependent variable. So in that case an event study is the poor sister of the experiment, but very useful in this research.
In this market model we follow the market, and if there is an outperformance of the market we find an abnormal return. The outperformance can also be analyzed in various windows. The results can be summed and calculated using the following market model.

Rit = ??i + ??i Rmt + Dikt + ??it , 'I,t.

The Rit is the return of a firm (i) on a certain day (t), the ??i is a constant ??i Rmt is the effect of the market model [MM], 'summation' stands for CAR realized after corrected for the MM, Dikt is the dummy variable which equals 1 for event days, ??it stand for the Error term (mu=0), and finally 'I,t stands for all firms (i) at a certain given time (t). Karafiath (1988) was the first introducing this model allowing a cumulative prediction error by including a dummy variable. Roscovan (2009) used 'a vector of dummy variables to the right-hand side of the corresponding equity market model'. Fields (2006) used a dummy variable for the renewal switch date because their first interest was to whether there was a change in abnormal returns once they accounted for possible changes in borrowers characteristics.

Data analysis
In table 1, we find for the full sample a two-day average excess return (t=0+1) of nearly 21,1 basis points. This is conducted out of the full sample, which contains 402 bank loan announcements. In studies conducted roughly from 1985-2000, larger results are found. For example, Mickelson and Partch (1986) found a two-day average positive abnormal effect of 89 basis point, James (1987) found an effect of 193 basis points and Slovin et al. (1992) found an excess return of 69 basis points over their full sample. Looking at these results we can state that in earlier studies the two-day average excess return was larger than we have found with our data. We then looked at different time periods in our data to examine whether there was any difference in the abnormal returns over these periods. The results reflected our expectations. We expected that during the period of the financial crisis (2007-2012) the cumulative abnormal return was larger than during the years before that. After all, banks didn't issue a lot of loans during the financial crisis. So when they did, they must have been sure of the potential and the credibility of the borrower. However, the differences were not of a reasonable impact after all. During the period 2007-2012, the two-day average excess return was almost 30 basis points over 40 bank loan announcements, which implicates that there was a positive difference of 10 basis points in comparison with the period 2000-2006. The two-day average excess return over 362 bank loan announcements then were 20 basis points. Our findings are consistent with Slovin et al. (1992). His findings are all positive and the CARs for the small firms are larger than the CARs for the large firms. Slovin et al. (1992) found a CAR of 192 bps for the small firms and a CAR of 48 bps for the large firms over the period 1980-1986.

TABLE 1
Yearly Mean Abnormal Returns (t=0:+1) of U.S. Based firms for all loans. The yearly Abnormal Returns are divided for the full sample, small, medium and large firms.
Full Sample Small Firms Medium Firms Large Firms

Year N CAR(%) N CAR(%) N CAR(%) N CAR(%)
2000 34 0,139 12 -0,311 14 0,301 8 0,531
2001 60 0,129 10 -0,00917 30 0,424 20 -0,243
2002 84 0,407 23 0,317 37 0,301 24 0,657
2003 91 -0,012 14 0,314 50 -0,115 27 -0,353
2004 32 0,167 7 0,606 15 0,093 10 -0,0305
2005 38 0,412 16 0,384 18 0,437 4 0,407
2006 23 0,712 10 0,793 11 0,765 2 0,013
2007 2 0,118 1 -0,0951 1 0,331 0 -
2008 7 0,193 2 -0,369 2 -0,25 3 0,862
2009 6 0,725 1 1,687 4 0,527 1 0,555
2010 3 -0,117 0 - 3 -0,117 0 -
2011 13 0,141 2 0,706 11 0,0388 0 -
2012 9 0,488 2 1,639 6 0,118 1 0,399
Cumulative two-day Abnormal Returns
Full Sample Small Firms Medium Firms Large Firms
Year N CAR(%) N CAR(%) N CAR(%) N CAR(%)
2000-2006 362 0,202 92 0,285 175 0,229 95 0,0733
2007-2012 40 0,295 8 0,692 27 0,101 5 0,708
2000-2012 402 0,211 100 0,317 202 0,216 100 0,105

In table 2 we described the summary of the descriptive statistics (mean, median, minimum and maximum) for the firm size (US$ Mil), the deal size (US$ Mil), the years to final maturity and the profitability for the 402 bank loan announcements between 2000 and 2012. We divided the full sample of 402 announcements into three categories; small, medium and large firms. By using SPSSstatistics we calculated the 25 and 75 percentiles from the total annual assets of the full sample to define the 'small', 'medium' and 'large' firms. All assets below the 25% boundary are defined as 'small' firms. All assets in between the 25% and 75% area is defined as medium sized. And all assets above the 75% boundary are defined as 'large' firms. On the next page, some
descriptive statistics of our sample are to be found.

TABLE 2
Summary of the descriptive statistics for Firm Size, Deal Size, Maturity and for Profitability categorized by firm size for 402 uncontaminated bank loan announcement over the period 2000-2012
Characteristic Mean Median Minimum Maximum
Full sample N=402
Firm Sizea 16672,162 6266,15 38,3 667543
Deal Sizeb 824,079 500 100 6000
Maturityc 3,492 3,71 0,167 12,008
Profitabilityd 0,223 0,0519 -0,419 21,910
Large firms N=100
Firm Size 52425,993 32928 16396 667543
Deal Size 1317,779 927,85 100 6000
Maturity 2,545 1,625 0,167 12,008
Profitability 0,684 0,0786 -0,199 21,910
Medium firms N=202
Firm Size 6679,209 6266,15 2159,6 16018,6
Deal Size 794,377 500 100 5000
Maturity 3,574 3,711 0,411 10,088
Profitability 0,806 0,0509 -0,419 1,213
Small firms N=100
Firm Size 1104,097 1034,05 38,3 2151,9
Deal Size 390,379 290 100 1400
Maturity 4,271 5,003 0,485 8,005
Profitability 0,0508 0,385 -0,1181 0,233
Notes: a Indicates Firm Size as total annual assets in million dollars. B Indicates the deal size in total million dollar. C Indicates the time to maturity in years. D Indicates the profitability measured as EBIT/total annual assets.
Remarkable are the massive growth of the loan package amount and the total assets of the firms over the years. Slovin et al. (1992) found an average of 201,5 million dollar loan agreement for large firms over the period 1980-1986. For the large firms we find an average loan agreement of 1317,779 million dollar, which means an increase of the average loan package of more then 600%.
The same remarkable development occurred for the total annual assets. The average years to final maturity declined strongly compared with the findings of Slovin et al. (1992).
IV. Regression Analysis

We will now use a multivariate regression to analyze the relation between the dependent variable, two-day cumulative abnormal return and the independent variables. We are using four independent variables in our regression analysis. The independent variables are defined as following. The first variable is Firm Size. This variable is measured in total annual assets. We measured the second variable, Deal Size, as the total value of the loan agreement. The third variable is the Maturity; the yield to maturity is measured in years. The last independent variable is the earnings before interest and taxes divided by the total annual assets of the corresponding company, given as Profitability.

Regression Analysis models
Model 1: Full sample CARI,t = ??I,t + ??1LnFSI,t+ ??2LnDSI,t + ??3YTMI,t + ??4EBIT/TAI,t + ??I,t
Model 2: Large firms CARI,t = ??I,t + ??1LnFSI,t+ ??2LnDSI,t + ??3YTMI,t + ??4EBIT/TAI,t + ??I,t
Model 3: Medium firms CARI,t = ??I,t + ??1LnFSI,t+ ??2LnDSI,t + ??3YTMI,t + ??4EBIT/TAI,t + ??I,t
Model 4: Small firms CARI,t = ??I,t + ??1LnFSI,t+ ??2LnDSI,t + ??3YTMI,t + ??4EBIT/TAI,t + ??I,t

We will now evaluate the separate expected effects of the independent variables on the dependent variable. Firm size, calculated as the total annual assets in US$ Mil, will have a large impact on the dependent variable CAR. We think when a firm is 'large', it will have less effect on the CAR, according to the results Slovin et al. (1992) provided. They argued that there was a different effect on the CAR between small and large firms. With large firms there was no evidence of any influence on the CAR. But when looking at the effect of small firms, they found a significant positive effect. This was also in line with Diamond (1984) who argued that the monitoring and screening services of the bank became less critical as the firm became larger. With respect to the second variable, the deal size, we believe it has a positive effect on the dependent variable CAR. This means that the larger the amount of the loan is, the larger the effect will be on the CAR. When we look at small firms, a large loan can be seen as risky by investors and therefore the CAR can be lower when compared with large companies (large loans are proportionally smaller for large firms). By risky we mean the ability of the company to repay the loan. The variable Maturity tells us how long the firm has to refinance their loan. A short YTM can give two signals to investors. One signal is that the bank trusts the firm that it is able to repay its debt in a short amount of time, on the other hand the short period of time makes the loan more risky, which could be a negative thing in the eyes of an investor. The last variable we use in our regression analysis is EBIT corrected for Total Assets. We believe that this independent variable also influences the dependent variable CAR. EBIT gives us more information about the profitability of the firm that borrows money. When a firm has a large EBIT, it is more likely that they will use internal investments instead of making use of bank loans. Therefore we think that when a company has a large EBIT the stock market's reaction to a bank loan announcement will have an adverse effect.
In all the four regression models we use a natural logarithm for the firm size and the deal size. The first model is related to the full sample. Model 2,3 and 4 are respectively related to the small, medium and large firms. The models can be found at the top of this chapter.
In table 3 there is a summary of the regression analysis for the CAR to be found.

TABLE 3
Summary of the regression analysis for CAR window (0,1). Independent variables: total annual assets, loan package amount, years to final maturity and earnings before interest and taxes. Categorized by firm size for 402 uncontaminated bank loan announcements over the period 2000-2012.
Characteristics Full
Sample Small
Firms Medium
Firms Large
Firms
Ln (Firm Size) -0,081 0,051 -0,113 -0,02
Ln (Deal Size) 0,139 -0,138 0,193 0,158
Maturity -0,006 0,049 -0,026 -0,017
Profitability 0,046 0,063 -0,474 -5,369
R2 0,015 0,021 0,019 0,078
N 402 100 202 100

The models show us that when medium and large firms are growing bigger this will have a negative effect on the stock price (-0,113 & -0,020) and that small firms will have a positive reaction on firm size (0,051). These outcomes are consistent with our expectations and also according to the statement from Diamond (1984) as mentioned above. Our findings contradict with the findings of Slovin (1992) for the firm size. We both found a negative response on firm size for the full sample, but for the small and large firms we find the opposite values. We find a positive correlation for small firms and a negative correlation for large firms. Slovin (1992) found -0,0192 for small firms and 0,0027 for large firms. In the model for the full sample we find a 0,139 coefficient for deal size, which means that when the loan package amount increases the stock prices react positive. The same thing occurs for the medium and large firms, except for the small firms. We expected this because the bigger the deal size is, the riskier this will become for small firms to repay the debt. That is why the coefficient is -0,138. For the small firms we see that the coefficient is positive (0,049). It tells us that the larger the time period is to repay the debt, the larger the positive effect on the stock price will be. For the medium and large firms this will have a negative effect on the stock prices. Slovin (1992) also finds that small firm reacts positive on maturity (0,0214) and also that large firms react negative on the length of maturity (-0,0048). The explanation for this might be that when these medium and large firms need more time to repay the debt, this will give the investors a signal that they are not able to comply with the requirements at a short term, which can be interpreted as a bad sign. Also the profitability shows us that when medium or large firms profit it will have a negative effect on the stock price. This might be explained by reasoning that they don't need a loan the finance activities when they make a profit and so taking an unnecessary risk. V. Meta Analysis

V. Meta Analysis

We completed the forest plot (figure 2) on the next page with the results of our own findings. The 9th study represents the results of our own research. Study 9 contains the full sample of the years 2000-2012. The blue dots in the plot represent the two-day cumulative abnormal returns of the studies. The red dot indicates the confidence interval and the prediction interval of all studies combined. All our own findings of the two-day cumulative abnormal return are smaller than all the studies we have used in our literature review. It is noticeable that the results of the confidence interval of our full sample do not fit in the range of the confidence interval of all studies combined.
We did find only positive cumulative abnormal returns; this means that a bank loan announcement has an average positive effect on the stock price of a company. This is important for firms that are thinking of strategies on how to finance their businesses due to the fact that their stock price will probably increase after a bank loan announcement has been made. Only a small increase of the stock price can result in a large increase of the total value of one company. However, as we can see in the range of the prediction interval, it is also possible that the result of a bank loan announcement results in a negative cumulative abnormal return, but as we can see that will happen less often than a positive cumulative abnormal return. It is new for us that the prediction interval on this subject can be negative for companies. We can see that by adding our own results to the Meta analysis, which have an overall lower effect size than previous studies, the prediction interval can be negative. This is very important for managers who are thinking of funding their firms' activities, because on basis of this study, a positive result is not necessarily very plausible anymore. When there is a risk of getting a negative cumulative abnormal return, managers will think of other investment options like for example obligations or private debt. Especially for large companies this can be more interesting due to the current information symmetry for publicly listed firms. When a large company has to lend money from a bank to fund their activity, it can be seen as a negative signal by shareholders and can therefor lead to a negative CAR.
The CAR for our full sample is 0,211% that is lower than the average of all studies that were held in earlier years. When we look at newer studies we find mixed results, with also an average lower CAR, but also with a CAR that is in line with earlier studies like Mickelson et al. (1986). Fields (2006) found a CAR of 0,13% for the period 2000-2003, Ross (2010) however found a CAR of 1,03% between the same years as Fields did, 2000-2003. This difference in outcome can be explained by the fact that the researchers used different databases. Looking at all the studies we cannot say that the CAR is decreasing over time, however Fields (2006) finds over the period 1980-2003, a decreasing effect in the cumulative abnormal return.
As we expected, the CAR for the years 2007-2012 is larger than the CAR for the years 2000-2006. This is most likely due to the financial crisis in the past years. For shareholders it was a sign of trust when a bank in this period gave a loan to a firm, hence the stock prices went up which resulted in a higher CAR. However this difference is almost negligible, for the period 2000-2006 the CAR was 0,202% and for the 2007-2012 period the CAR was 0,295%. However, as stated before, little difference like this can have a relatively large economic impact on a firm. Yet, there has to be noted that the average CAR is decreasing compared to earlier studies, and this means it will have less economic impact on a firm's value and on the economy as a whole.
We think the reason why the overall CAR from our study is lower than al the other studies are due to the information symmetry in nowadays world. Everybody, including investors, banks, potential clients and so on, have more access to any information than ever, which leads to the fact that every player in the financial field can, and will, be aware of the creditworthiness and reasons of a firm that is borrowing money or of the bank that is lending it. This means that the monitoring and screening services from the bank no longer have an effect on the stock price because every party is able to do this. Not only due to the Internet, but also due to the obligation a listed firm has to publish his books. The competitive advantage for banks is gone, and so is the competitive advantage for the company's stock price. This is in line with the results of Fields (2006) who stated, 'we find strong evidence to suggest that bank loan announcement returns have become insignificant in the years following studies by James (1987) and Lummer and McConnell (1989)'. However, the CAR is still positive which must mean that not everything is known at other (public) parties or that there are more influences on the stock price reaction to a bank loan announcement. We expect both. Banks still have a little advantage over other financial intermediaries and there are of course more influences on the stock price reactions than just information asymmetry.


#: Study: N:
1 Mikkelson et al. (1986) 360
2 James (1987) 80
3 Lummer and McConnel (1989) 728
4 Slovin et al (1992) 676
5 Billet et al. (1995) 540
6 Aintablian and Roberts (2000) 122
7 Fiels et al. (2006) 1111
8 Ross (2010) 1064
9 Own Study (2000-2012) 402

VI. Conclusion

Let us in the first place make something very clear. After all, much can be stated on the subject of stock market reactions to bank loan announcements, but one thing remains very certain: it is certainly not a very precise topic where robust conclusions are able to be drawn on. We have discovered that there are so many factors that all have to be in the exact same setting in order to say something realistic, to compare in the right way and try to explain development through time. All studies that are conducted in the past and included in this thesis, differ at least in some way from each other. This leads to great differences in results, even when the same time period is reviewed. The best example to be found in this thesis is the difference between the studies of Ross (2010) and Fields (2006), which have a great difference in CARs for the same time period of 2000-2003. From here, we will conclude our study but the statement as just made has to be kept in mind.
In this study, some very remarkable results came out of our data set. We found that when the global financial crisis started at the beginning of 2007, very few loans were issued. Of course this is logical, for most banks started to deal with liquidity problems and were hardly able to issue loans from that point. In our data, 362 loans were issued before the crisis, and 40 after. Most loans that were still issued were issued at medium-sized firms. One would expect CARs to be higher during the financial crisis for the reasoning that when a firm gets a loan when loans are hardly issued, the issuer must have a very great confidence in the borrowing firm. This expectation is somewhat supported by our results, but the difference is so small (0,093%) that we consider it negligible. A possible explanation for this could be that stock markets were so unpopular at that time that even when a firm was expected to perform well, stock buyers were extremely careful to buy stocks. This however is hard to prove and supports our first statement that this topic is very hard to draw robust conclusions on.
Also remarkable is the fact that CARs seem to remain positive on average. Our own results made the prediction interval partly negative but just for a negligible amount. Our results combined with the results of most included studies however do suggest that CARs have become slightly smaller roughly from 2000. This can be explained by the fact that information asymmetry has become significantly smaller from roughly the start of the 21st century. This of course because of the Internet and regulations demanding information of publicly listed firms to be reported openly. The question whether bank loan announcements still do matter is a very obvious one and we can conclude that nowadays the CARs thus still are positive, but almost have a negligible 'probably economical and maybe even managerial irrelevant- effect. We can imagine that due to completely other types of announcements CARs are very different and perhaps much greater. That would be a interesting topic for future studies.
In our study, most of our expectations and possible explanations are reflected in the results. We find, in line with for example Fama (1985) and Diamond (1984), that Firm size is negatively correlated with CARs. Small firms thus seem to cope in a greater way with moral hazard and information asymmetry. We also find that when a firm is small, the deal size is negatively correlated with CARs. This might be explained by the fact that when a loan is larger, this is relatively more risky when compared to a large firm. We can also see that maturity has a positive effect on CARs at small firms and a negative effect at large firms. We can imagine that large firms are more expected to be able to pay their debts quickly then small firms, and that therefor when YTM is large at a large firm this is a negative sign of liquidity. The last effect we mention is the extremely strong and convincing negative effect that is found at profitability at large firms. This very strong effect indicates that when large firms are very profitable, it would be strange to externally finance activities. This strange sign is obviously interpreted as a strong negative sign of future performance. This probably is the most robust effect we have found in our study.
Overall, we can say that our results are pretty much in line with current literature on everything that is discussed. Firm size really seems to matter, maybe especially nowadays, in CARs. The CARs themselves however seem to have shrunk and thus it is the question whether this difference between these already small CARs are worthy to take into account at all. A far more interesting and strong recommendation we unintentionally make is that large firms really have to be careful with obtaining loans when they are relatively profitable.
In the end, we question whether CARs after bank loan announcements are still worthy to examine. All studies and our own study however, raise questions that are extremely interesting to examine in future studies. For instance: What are the exact differences between all different information sources? What would be the ideal setting to study CARs after bank loan announcements (for now every study is different and thus slightly incomparable)? What types of announcements are of great economical and managerial relevance nowadays? Are other foreign financial markets becoming more comparable to the US stock markets due to globalization and financial integration, and are thus the same effects to be found there? These questions are extremely interesting and we think that robust results of studies examining these questions might be very relevant.

VII. References

Becketti, Sean and Morris, Charles (1992) 'Are bank loans still special'?, Economic Review Federal Reserve Bank of Kansas City, 77, 3: 71-84.
Berger, A. (1999). The 'Big Picture' of relationship 'nance,in 'Business Access to Capital and
Credit' (J. L. Blanton, A. Williams, and S. L. Rhine, Eds.), pp. 390'400. A Federal Reserve System Research Conference.
Berger, A., and Udell, G. F. (1995). Relationship lending and lines of credit in small 'rm 'nance, J.
Business 68, 351'381
Best, Ronald and Zhang, Hang (1993) 'Alternative information sources and the information content of bank loans', The Journal of Finance, 48, 4: pp. 1507-1522.
Billet, Matthew T., Flannery, Mark J. and Garfinkel, Jon A. (1995) 'The effect of lender identity on a borrowing firm's equity return', The Journal of Finance, 50, 2: pp. 699-718.
Carter, Richard and Manaster, Steven (1990) 'Initial public offerings and underwriter reputation',
The Journal of Finance, 45, 4: pp. 1045-1067.
Coleman, Anthony D.F., Esho, Neil and Sharpe, Ian G. (2006) 'Does bank monitoring influence loan contract terms'?, Journal of financial service research, vol. 30: pp. 177-198.
Collins, Daniel W., Korthari, S.P. and Dawson, Rayburn, Judy D. (1987) 'Firm size and the information content of prices with respect to earnings', Journal of Accounting and Economics, vol. 9: pp. 111-138
Diamond, D. (1984) 'Financial intermediation and delegated monitoring', Review of Economic Studies, vol. 51: pp. 393-414.
Diamond, D. (1991) 'Monitoring and reputation: The choice between bank loans and directly placed debt', Journal of Political Economy, vol. 99: pp. 689-721.
Fama, Eugene F. (1985) 'What's different about banks'?, Journal of Monetary Economics, vol. 15: pp. 29-39.
Fields, L. P., Berry, Tammy L., Byers, Steven and Fraser, Donald R. (2006) 'Do bank loan relationships still matter'?, Journal of Money, Credit and Banking, 38, 5: pp. 1195-1209.
Freeman, R. N. (1986) 'The association between accounting earnings and security returns for large and small firms', Journal of Accounting and Economics, vol. 9: pp.195-228.
Gonzales, Laura A. (2011) 'Dogs that bark: Why are bank loan announcements newsworthy'?, Global Economy and Finance Journal, 4, 1: pp. 62-79.
Hadlock, Charles and James, Chris (2002) 'Do banks provide financial slack'?, Journal of Finance, vol. 57: pp. 1383-1419.
James, Christopher (1987) 'Some evidence on the uniqueness of bank loans', Journal of Financial Economics, vol. 19: pp. 217-235.
Kysucky, Vlado and Norden, Lars (2013) 'The Benefits of Relationship Lending in a
Cross- Country Context: A Meta-Analysis', Rotterdam School of Management,
Erasmus University
Karafiath, I., 1988, 'Using Dummy Variables in the Event Methodology', The Financial Review - vol. 23, No. 3, pp. 351.
Lee, Kwang-Won and Sharpe, Ian G. (2009) 'Does a bank's loan screening and monitoring matter'?, Journal of financial services research, vol. 35: pp. 33-52.
Levine, Ross, and Zervos, Sara (1996) 'Stock market development and long-run growth', The
World Bank Economic Review, 10, 2: pp. 323-339.
Lummer, Scott L. and McConnell, John J. (1989) 'Further evidence on the bank lending process
and the capital-market response to bank loan', Journal of Financial Economics, vol. 25: pp. 99-122.
Mikkelson, Wayne H. and Partch, Megan M. (1986) 'Valuation effects of securities offerings and the issuance process', Journal of Financial Economics, vol. 15: pp. 31-60.
Ross, David G. (2010) 'The 'Dominant Bank Effect:' How High Lender Reputation Affects the Information Content and Terms of Bank Loans', Review of Financial Studies, 23, 7: pp. 2730-2756.
Slovin, Myron B., Sushka, M. and Hudson, C. (1988) 'Corporate commercial paper, note issuance facilities and shareholder wealth, Journal of International Money and Finance, vol. 7: pp. 289-302.
Slovin, Myron B., Glascock, John L. and Johnson, Shane A. (1992) 'Firm size and the information content of bank loan announcements', Journal of Banking and Finance, vol. 16: pp. 1057-1071.

VIII. Appendix
Study: Authors: Year of Publication: Research Strategy: Sample Description: Effect Sizes (in base points): Standard Error Calculation: Confidence Interval:
"The 'Dominant Bank Effect:' How High Lender Reputation Affects the Information Content and Terms of Bank Loans" David Gaddis Ross 2010 Event study Source: Factiva, Compuscat, CRSP 1064 loan announcements period: 2000 - 2003 Average excess return: 103 bps Z=6,20 Effect=1,03% SE=0,19/6,20=0,03064 0,704%-1,356%
'Bank and Bonds: The Impact of Bank Loan Announcements on Bond and Equity Prices'' Steven Ongena Viorel Roscovan Wei-ling Song Bas J. M. Werker 2008 Event study Source: Datastream and LPC 3589 bonds issued (1997:2004) 896 loan deals (1997:2003) involving 364 U.S. firms. Credit Spread(-1;+1) (bonds): -16,86 bps No idea on how to? ?
"Do bank loan relationships still matter''? L. Paige Fields, Donald R. Fraser, Tammy L. Berry, Steven Byers 2006 Event study Source: Lexis/Nexis 1111 bank loan announcement samples publicly trading U.S. firms Period: 1980-2003 Average excess return: 46 bps (1980-89): Effect size= 0,60 SE= 3,06% (1990-99): Effect size= 0,51 SE=3,38% (2000-03): Effect size= 0,13 SE= 3,51% (1980-89): 0,540%-0,660% (1990-99): 0,444%-0,576% (2000-03): 0,061%-0,199%
'A note on market response to corporate loan announcements in Canada''. Sebouh Aintablian, Gordon S. Roberts 2000 Event study Source: Canadian Corporate News and Canada Newswire 122 bank loan announcement samples 14 industries Period: 1988-1995 Average excess return: 122,56 bps T=5,62 Effect=1,2256% Mu=0 SE=(1,2256-0)/5,62 = 0,2181 0,794%-1,657%
"Stock market development and long-run growth" Ross Levine and Sara Zervos 1996 Event study Not yet found out Not yet calculated Not yet calculated Not yet calculated
'The effect of lender identity on a borrowing firm's equity return' Billett, M, M Flannery, and J Garfinkel 1995 Event study Source: Dow Jones News Retrieval Service. 626 Bank loans announcements. Period: 1980 - 1989 Average excess return: 63 bps t=3,625. Effect=0,628%. Beta0=0. SE: (0,628-0)/3,625=0,173 0,288%-0,968%
"Firm size and the information content of bank loan announcements" Myron B. Slovin, Shane A. Johnson, John L. Glascock 1992 Event study 676 announcements of credit agreements between firms on NYSE and NASDAQ in the period 1980-1986 Average excess return: 69 bps Z=4,23. Effect=0,69%. SE=0,69/4,23 = 0,163 0,370%-1,010%
"Monitoring and Reputation: the choice between bank loans and directly placed debt" Diamond, D.W. 1991 Theoretical Paper N/a N/a N/a N/a
"Further evidence on the bank lending process and the capital-market response to bank loan agreements" Scott L. Lummer and John J. McConnel 1989 Event study 728 credit agreements between U.S. corporations with stock prices on the CRSP daily file of NYSE and AMEX and U.S. or foreign banks for the period 1976-1986. Average excess return: 61 bps Z=2,69. Effect=0.61% 0.61/2.69 = 0.227 =SE 0,165%-1,055%
"Some evidence on the uniqueness of bank loans" James, C 1987 Event study Source: Center for Research on Security Prices. 300 random selected NSYE listed companies with announced debts. Period: 1974 - 1983 Average excess return: 193 bps Z=3,96. Effect=1,93%. Mu=0. SE: (1,93-0)/3,96=0,487 0,960%-2,900%
"Valuation effects of security offerings and the issuance process" Wayne, H. Mikkelson and Megan M. Partch 1986 Event study Source: 360 random selected NSYE listed companies with private credit agreements. Period: 1972 - 1982 Average excess returns after a credit agreement: 89 bps Z=2,58. Effect=0,89%. Mu=0. SE: (0,89-0)/2,58=0,345 0,212%-1,568%
Corporate financing and investment decisions when firms have information that investors do not have'' Steward C. Myers Nicholas S. Majluf 1986 Theoretical Paper N/a N/a N/a N/a
"What's different about banks?" Eugene F. Fama 1985 Theoretical Paper N/a N/a N/a N/a
"Financial intermediation and delegated monitoring" Diamond, D.W. 1984 Theoretical Paper N/a N/a N/a N/a

Source: Essay UK - http://ntechno.pro/free-essays/finance/stock-market-reactions-bank-loan.php


Not what you're looking for?

Search:


About this resource

This Finance essay was submitted to us by a student in order to help you with your studies.


Rating:

Rating  
No ratings yet!

  • Order a custom essay
  • Download this page
  • Print this page
  • Search again

Word count:

This page has approximately words.


Share:


Cite:

If you use part of this page in your own work, you need to provide a citation, as follows:

Essay UK, Stock Market Reactions To Bank Loan Announcements. Available from: <http://ntechno.pro/free-essays/finance/stock-market-reactions-bank-loan.php> [20-10-17].


More information:

If you are the original author of this content and no longer wish to have it published on our website then please click on the link below to request removal:


Essay and dissertation help

badges