Emmanuel Yimfor: Misconduct synergies | Gerald R. Ford School of Public Policy

Emmanuel Yimfor: Misconduct synergies

September 9, 2021 0:47:24
Kaltura Video

Professors Yimfor and Tookes use the investment advisory industry as a laboratory to test whether there are improvements in employee misconduct following M&A events ("misconduct synergies''). September, 2021.

Transcript:

0:00:05.7 Christie Baer: Welcome. My name is Christie Baer. I'm the assistant director of the Center on Finance, Law and Policy. Welcome to our first blue bag lunch talk of the 2021/22 academic year. These talks were founded when the Center was founded seven years ago. And the purpose of them is to take advantage of the strength of U of M where every sidewalk leads to a top 10 program, and allow faculty from different schools to present their ideas. And before a friendly, supportive, and yet wicked, smart and super supportive but helpful audience. So these talks have been happening for a really long time, the topics very super wild, wildly, widely, either way. And these are a chance for faculty to say what they're working on sometimes before they get to a published paper, sometimes when they just have a working paper. And so I hope that you'll do this as a chance to ask every question that occurs to you. If you read the paper already, fantastic. If you didn't, you don't even need to worry about it, because Professor Yimfor is gonna tell you all about it. And then you can go back and read it after.

0:01:19.3 CB: So our speaker today is Professor Emmanuel Yimfor from the Ross School of Business. He joined the Ross School during COVID. And so is meeting people in real life this year. So if you haven't seen him in person in the faculty lounge, and you're like, who is that guy? You should go and talk to him because he doesn't know what you look like in 3D. His research is focused on empirical corporate finance with a focus on financial intermediation, capital formation, entrepreneurial finance. And he's specifically interested in how information frictions affect the ability of startups to raise external financing. He is not teaching this semester. So if you're a student, and you were hoping to sign up for his class, you have to wait till next semester. Next semester though, he's all yours.

0:02:12.2 CB: He's going to be teaching venture capital, private equity, global private equity and entrepreneurial finance courses next semester. His PhD is from Rice. He has a master's from Kent State. And though he did spend a significant amount of time in Ohio, he is thrilled to be a Michigan Wolverine now. So his talk today is, he has a fancy title for it, but we're gonna think of it as employee misconduct and then an investment firm context, because we all want to hear about when employees behave badly. And so he will now tell you what his talk is actually about. He's gonna present for a little bit and at the end, everyone will have a chance to just ask questions either in the chat, you can send them to everyone in the chat, or you can raise your hand at the end. And so, from here, let me turn it over to Professor Yimfor, thanks for being with us today.

0:03:19.5 Emmanuel Yimfor: Thank you for the wonderful introduction, Christy. Let me get my slides to full screen mode. Can you see them? Alright, excellent. So today I'm gonna be discussing new work with Heather Tookes. She's at Yale. And in this paper, what we're looking at is the relationship between corporate control transactions and employee behavior. Now in finance, we're very interested in allocating capital to it's most productive use. Now, that's no different when it comes to mergers and acquisitions. As far as investments go, mergers are one of the largest investments that a firm would ever make. And so it's very natural that a lot of academic papers have kind of asked the question, are mergers a profitable investment? 

0:04:09.1 EY: Now how have these papers done that? Well, they have looked at various things, for issuers, for firms that are publicly traded, they've looked at things like announcement period returns, post takeover stock returns, changes in profitability, but there's really no evidence on the exact precise mechanisms through which the synergies are realized. We're gonna take a different approach and we're gonna get into the weeds here and ask, can M&A, this corporate control transactions, can they improve the behavior of rank and file employees. Now, the investment advisory industry is a very useful laboratory because as I would show you, registration and licensing requires that the information about investment advisory firms and more specifically investment advisors be the public.

0:05:00.0 EY: Now, I'm gonna talk a lot about employee behavior. I'm gonna talk a lot about misconduct, sometimes I'm gonna call them disclosures. Let me give you a little bit of a feel for what these are. So I grabbed an example. So here the New Jersey bureau of securities here something that I'll call a regulatory disclosure. The New Jersey Bureau securities brought an action against Saviano, I'm gonna be calling him an advisor, an investor advisor. And they allege that he engaged in dishonest or unethical behavior in relation to borrowing money from his clients. Right? And for this they fine him about $20,000. So here's an example of a regulatory disclosure. Another example maybe is a customer dispute. So here the customer was unhappy with the kind of investments that the advisor purchased for the customer. They complained about it, and this particular dispute ended up being settled for about $45,000. Regulatory actions and customer disputes, those are the most popular categories of disclosures that constitute essentially what I'm gonna also be calling misconduct, and I will be a little bit more specific about why I will be calling some of these disclosures misconduct in a few slides.

0:06:17.0 EY: Now you should know, there are about 7% of all brokers, so these are individual advisors employed between 2004 and 2019, have at least one such disclosure. And a paper is not gonna be the first to talk about these disclosures, right? The paper is about corporate control transactions, M&A, and the effect they have on these disclosures, but what have people kinda looked at in the past, so this very popular paper, Egan, Matvos, and Seru, published in the JPE showed that this conduct is widespread, one in 13 advisors has a misconduct disclosure and to show that they are costly, the mean fine, so the fine is like I showed you a fine for the two examples that we looked at earlier, the average fine, if you look at the cross-section of these disclosures is about half a million dollars and the median settlement is about $40,000.

0:07:07.6 EY: Now, here's the fact that I want... We'll come back to this, but I want you to hold that in your minds for now, okay. One third of these advisors with misconduct are actually a repeat offenders, so if I know that you committed misconduct today, there's a very high chance that I know that you're gonna repeat it in the future, given that a lot of these advisors with misconduct are repeat offenders. Now, we also know that misconduct is contagious, in a paper by Dimmock, Gerken & co authored, what they show is that when advisory firms merge, for example, and a co-worker that doesn't have a misconduct record is exposed to a new co-worker that has a misconduct record, that employee is more likely to commit misconduct in the next three years and so the interpretation there is that fraud is contagious, right? 

0:07:55.9 EY: And so, A, this misconduct events are costly, B, this misconduct events are contiguous. If we zoom out a little bit, what is the consequence of this misconduct within the disclosures. On the investment advisory industry in general, there's a really cool paper by Gory Stuffman and Junker that show that actually these things can be very costly to the investment advisory industry, they use the Bernie Madoff... Bernie Madoff by the way, own an investment advisory firm, and it looked at how did his clients react when he was revealed as being a fraudster and it show that a lot of those clients were... Withdrew their assets from investment advisers, so even those that were not directly affected, but those for example that knew people who were directly affected withdrew their money from investment advisors and instead took that money to what they believe was safety, to banks.

0:08:55.5 EY: And so you know, these disclosures events are, A, costly, and B, contagious, and C, they can have broad-reaching impacts on the investment advisory industry in general. And so with that at the back of our mind, again, going to our specific question of how did this corporate control transactions affect employee behavior, we're gonna frame our test through the lens of existing theory, what do I mean by that? When you look back at the theory and the theory kinda suggests which type of targets and acquirers would merge with each other, there's a broad theory that I title, "The Market Discipline Hypothesis." Right? That suggests that better behaved firms will buy poorly performing firms, and there's a lot of empirical support for this idea, so for example, people have found that firms that have high market-to-book, if you don't know what market-to-book is, just think about that as better firms, tend to buy firms that have low market, to book maybe poor performing firms. And so then... I'm gonna get in a second too what this specifically implies for our specific context, which is mergers in the investment advisory industry and business conduct related disclosures that we've been talking about. But just hold that.

0:10:10.0 EY: Market discipline hypothesis, better behaved firms would buy poorly performing firms. Now, the compliments hypothesis suggests instead, that you would have a like bias like in mergers and acquisition, so if two firms maybe have similar cultures, those two firms are more easy to integrate post a merger and it's easy to realize this synergy following the merger, and so you really have two completely empirical predictions about who buys who. Now what do this hypothesis imply for our context, again, which is corporate control transactions in the investment advisory industry as a function of employee behavior of either the target or the acquirer. Well, the market discipline hypothesis actually has some pretty specific predictions. Number one, we would predict that high misconduct firms would be more likely to be targets than low misconduct firms; the better behaved ones, will be more likely to be acquirers. Two, we would predict in terms of merging, that low misconduct acquirers, again, who are better behaved, would merge with high misconduct targets, so again... And if you are a little bit... You can come closer to your screens, what's the difference between the first and the second, the first one is about who gets to enter the market for mergers and acquisitions.

0:11:31.0 EY: The second one is conditional on entry, how do they merge. And the third one says any post-merger reductions in misconduct to the extent that they are value relevant will be driven by changes related to the target from employees, because again, remember, according to the market discipline hypothesis, the target from employees are not well-behaved. And that's one of the reasons why they enter into the M&A market in the first place. What is the complements hypothesis imply? Well there's no real prediction about who gets to enter the market for mergers and acquisition, because it just suggests that firms with similar culture, firms that are alike would merge with each other. I'm invariable of interest as employee misconduct so there's a specific prediction about how they merge conditional on entry. Targets and acquirers should merge according to levels of employee wrong doing. And the compliments hypothesis also suggests that if there is any post-merger reductions in misconduct it could be driven by changes related to the target from employee, they're acquiring from employee or maybe even both.

0:12:42.8 EY: Let me tell you very briefly about our data. So the first data set, I kind of briefly discussed, we need the disclosure records of each of these advisors. So we wrote the script to collect that data for about 1.2 million advisors I believe. The next thing we need obviously is data on mergers. And so, we're gonna collect that from a few different sources. I'm not gonna bore you with the details, but just so you know, we have 419 mergers in our sample from 2004 to 2020. You should also know that the target and the acquirer are US based investment advisors. All right. Now I keep talking about employee misconduct. I showed you a specific example. In general, there are about 23 different categories, right? The most frequent categories are regulatory disclosures, which I showed you and customer disputes that are settled or are awarded some kind of a judgment.

0:13:49.8 EY: And so you can see here that we could use all 23 categories of disclosure and call them misconduct. But as Egan, Matvos, and Seru have showed, some of these are not a mild disclosures, right? They don't warrant the title of misconduct necessarily. And so, they did a lot of the work for us and they separated these categories of disclosures into ones that are a little bit more severe, that would be more related to misconduct. And these are the six categories that I have over here. And so, I'm gonna really be using two, we are gonna be using two measures of this disclosures at the employee level. The all measure is gonna count the total number of these disclosures over the past two years. And the EMS measure short for Egan, Matvos, and Seru, is gonna count just the six categories that are more severe, the more severe categories essentially of disclosures. I should note that the authors Egan, Matvos, and Seru, they show that advisors that have committed any one of this other disclosures are also more likely to have committed a misconduct under their more rigorous definitions. So we're gonna interpret changes in either the all measure, which encompasses all the 23 categories, or just the more strict categories that Egan, Matvos, and Seru think are more egregious as misconduct.

0:15:20.4 EY: Now I told you we had 419 mergers here. And so here, we're looking at some statistics of targets and acquirers that are in our sample. Now, I don't think it's surprising. Excuse me. I don't think it's surprising that on average targets tend to be a lot larger than acquirers. They tend to have a lot more assets under management, right? Now related to the hypothesis, one thing that's really interesting here is that you can already kind of see some evidence against the market discipline hypothesis, right? You can see here that targets are not better behaved necessarily relative to acquirers, set that backwards. Acquirers are not better behaved relative to targets. You can see that there's almost no difference in the employee disclosures of acquirers and targets. If anything, acquirers have experienced more recent growth in employee disclosures relative to targets. And so this is already a little bit of evidence against the market discipline hypothesis of why firms merge, and maybe this kind of suggests the compliment hypothesis of light bias light, but I'm gonna do a lot more digging in the next few slides.

0:16:38.5 EY: Now before we get into, okay, who gets to enter the M&A industry as a function of their misconduct history, how do they merge, and is there any drop in misconduct following the merger event. Let me first try to convince you, or just suggest the value relevance of misconduct, and how exactly are we gonna do that? We're gonna do that by looking at the relationship between employee misconduct, you're gonna remember that this is just, if you have 100 employees, 10 of them have this misconduct related disclosures in the last two years, for example, then that variable is gonna be 10%. That's exactly how we're measuring misconduct. A unit of observation here is an investment advisor here, right? We're gonna be looking at the cross section of investment advisors, and we're gonna ask if you have more of this employee misconduct, does that predict your level of assets under management, your change in assets under management, or whether or not you fail in the future? 

0:17:38.2 EY: And what we find is if you have this misconduct related disclosures, you have lower future assets under management. And so here, the estimates imply that a one standard deviation increase in this employee disclosures is associated with 8.3% lower future assets under management, right? And so, again, I'm not making a causal statement here. I'm just trying to convince you by showing you some co-relations, that misconduct is value relevant. If we look at change in assets under management, so here we're looking at the recent change in misconduct, and how is that related to future changes in assets under management, we find the same effect. Relative to the unconditional mean here, the estimates here imply that once in the deviation change said employee disclosures is associated with between a 3.5 and a 4.5% decrease in future changes in assets under management. Now, if you're not familiar with the investment advisory industry, you should know that the way most of these advisors, the investment advisory firms earn their revenue is by charging a fraction of assets under management as a fee. So they charge 1% or 2% of assets under management as a fee. And so assets under management here is very value relevant.

0:18:55.0 EY: The next thing We'll look at is failures, so we say, for example, if you have this recent employee disclosures, are you more likely to fail? And here again, relative to the unconditional mean 'cause failure is very rare in the sample, you're anywhere from 5% to 7% more likely to close if you have more of these employee disclosures. And so, this is just to convince you that there is a relationship, misconduct can devalue relevant, even if you don't believe that misconduct is necessarily causing these variables to change, at the very least, misconduct is associated with something that is value relevant to the extent that that's true, maybe it is a variable that potential acquirers will focus on when they're doing the merger, okay.

0:19:40.3 EY: And so, the next thing we're gonna look at is the first test that I showed you when we were talking about the hypothesis which is, who gets to enter the investment advisory market as a function of their misconduct history? Now again, remember that the market discipline hypothesis would predict that if you're an acquirer and you have a high level of employee disclosures, you're more likely... If you are a target, you're more likely to have a high level of misconduct related disclosures. Your employees are not well-behaved and that's why people are trying to take you over so that they can realize the synergies by improving the behavior of your employees.

0:20:19.9 EY: On the other hand, the market discipline hypothesis would predict that as an acquirer, you're gonna have better-behaved employees and as such, your misconduct-related disclosures should be low. What do we find? Well, we don't find evidence that's consistent with the market discipline hypothesis. Here, the estimates imply that if you have a one standard deviation increase in employee disclosures relative to the unconditional mean, you're about 12% less likely to be a target for an acquisition again, which is not consistent with the market discipline hypothesis where we expect these signs to be the opposite of what they are. When we look at the target, the relationship is the same, although when we do put control variables it's not as significant.

0:21:06.7 EY: And so here, when we're looking at entry into the market for M&As, what we conclude really is that the evidence is not consistent with the market discipline hypothesis. It seems to point more towards potential complementaries in misconduct, especially if you group this with the statistics that I showed you at the start there where we just look back, what does employee disclosures for the target look like? What do employees disclosures for the acquirer look like? They didn't seem to be that different from each other, if anything, target seemed to have recent growth in employee disclosures. And so what we're gonna do next is we're gonna test for assortative matching on these employee disclosures. Is it true that we really have a like-buys-like work where firms with high levels of employee disclosure are more likely to match with other firms with the same level of employee disclosures.

0:21:57.7 EY: Now, how are we gonna do that? I'm gonna walk you through it because it could be counter-intuitive when you first hear about it. So what we're doing now is, is there matching on employee misconduct in M&A? Now to do that, let's start with a very simple example, so here are two mergers in our sample in 2015. So Washington acquired Halsey and Pinnacle acquired Enrichment. Now, how are we gonna do the counter-factual mergers? In an ideal world, we would like to see all the targets that Washington considered. We would like to see all the targets that Pinnacle considered. We will ask, "How does Halsey differ for example, from all the other targets that Washington considered? Well, what empiricist and we don't observe that counterfactual.

0:22:50.8 EY: So we're gonna approximate it by doing the following. We're gonna take Halsey and we're gonna pair Halsey with Washington and Pinnacle, we'll do the same thing for Enrichment. And so, two of these mergers actually happened and two of them did not. Now we can observe employee disclosures for Halsey and for all the firms here, and so what we're gonna do is we're gonna create four buckets. We're gonna create... Our first bucket is where the acquirer and the target both have this misconduct-related disclosures. The second bucket, the target has misconduct-related disclosures but the acquirer doesn't. The third bucket, neither of them do, and the fourth bucket, only the target has these misconduct-related disclosures. And we're gonna ask, in each of these buckets, what fraction of these firms are actually true merging pairs to test for evidence of matching and misconduct? 

0:23:46.4 EY: So here's what our figure looks like. So again, recall that here for example, this tall bar would be the acquirer has some type of disclosure and the target has some type of disclosure, and so we see here that 4.15%of all pairs in this bucket are true merging firms. When we create this counterfactual pairs, the unconditional merger rate is about 2.25%. So this is 83% larger, and so when we look at target and acquirer pairs with no disclosure, we also see that that's higher. And so the evidence here does seem to suggest matching on misconduct. We could be a little bit more rigorous. How? We could actually run a regression where we can hold different things fixed.

0:24:31.5 EY: So I showed you the counterfactual pairing. Some of them are fake mergers, others are real. What we're gonna ask is, if the distance in disclosure between the acquirer and the target is high, and we have assortative matching on misconduct, then we would expect that to be negatively related to the probability of a merger. We're gonna hold other characteristics of the target and acquirer fits. So how should do interpret retail clients here for example? Retail clients would be an indicator that equals one if both the target and the acquirer both serve retail clients. We're gonna do the same thing for the other control variables. Now, what we find there again is evidence of assortative matching on misconduct. So here, a standard deviation increase for example, in the distance in employee disclosures between the acquirer and the target, relative to the unconditional mean here is associated with about a 20% decrease in the probability of a merger which does support the like-buys-like hypothesis, the complement hypothesis of mergers in the investment advisory industry.

0:25:42.2 EY: Now, the next thing we're gonna move to is does misconduct dropped following the merger? Is there evidence what we're gonna be calling, is there evidence of misconduct synergies? And this misconduct synergies I should come back to this again, I recall that I had shown you some suggested evidence that this misconduct related disclosures are value relevant, right? If you had higher levels of employee misconduct, you are more likely to close, which is costly, not just to the firm, but to the firm's clients. If you had a higher level of employee disclosures, your future assets and their management were likely to be low and even the change in assets and the management as a function of a change in employee misconduct, those things, were negatively related. And so does misconduct drop falling the merger? Now, before I show you our finding, I wanna set this stage a little bit by showing you a little bit of the detail.

0:26:34.0 EY: Remember Halsey and Washington from earlier, they merged in 2015. Now, to conduct this test, we're gonna assume for example, that they merged in 2012 and then we're gonna track the merger all the way to 2018. So three years before the merger, three years after the merger. Of course, we can observe the disclosure for Halsey up until the merger when it gets absorbed by Washington, we can observe the disclosure for Washington. And all we're gonna do is we're gonna say, "What is the weighted average of employee misconduct for the target? And the acquirer say in 2012, and this is my simple example here, "They had an equal number of employees so the average is gonna be 1.5." Obviously after the merger, we're just gonna be tracking the combined firm here and so the target is now gonna have a value for disclosures. If you have questions about this we can always get back to it. And so, what do we find? What we find, a drop in misconduct following the merger by anywhere from 25%-34%, depending on the measure of misconduct that you use. There's also a decline in the growth of recent misconduct following the merger and in the paper, we do a lot of gymnastics essentially to make sure that this result is robust. And so, if you have some questions about that, we can come back to it or you can take a look on my, at the paper, which is, I believe posted on my website.

0:28:05.5 EY: Now misconduct falls following the merger, is that an artifact of the way we constructed the sample. In order to find out whether the drop in misconduct is actually as a result of the merger, we need to go after the mechanism. Why does misconduct essentially fall following the merger? Right, now there are few reasons why misconduct might fall following the merger. It could be the case that existing employees of the combined target and acquiring firm suddenly start to behave better. "Hey, after the merger, I met this other guy who turns out, does something that's very similar to me. Lunch was all nice and fun, but they might lay me off and keep him instead. I better get my act together," and so everybody start behaving better. Or it could be the case that after the merger the combined firm, they let go suddenly of a lot of employees that have this misconduct related disclosures and so we're gonna call that the separation hypothesis and that's what we're gonna test next. Here in this set, new set of tests, a unit of observation is gonna be an employee, an individual advisor working for either the target or the acquirer in the five years before the merger.

0:29:20.6 EY: We're gonna simply ask, "If you have misconduct related disclosures, recent instances of misconduct, are you more likely to get fired after the merger?" Post is gonna be an indicator equals one for the years following the merger. And so, what we find is indeed you are more likely to be fired following the merger. If we look... So here are the results for target firm employees only. If we look for example at the sensitivity of these disclosures to separation before the merger happens, we see that they are not statistically significant, but suddenly after the merger happens, employees with this disclosures are more likely to be fired irrespective of the measure of misconduct that we use.

0:30:09.1 EY: Now, one way to interpret this result is, the target firm manager, for example, had maybe close relationships with some of these employees. He thought maybe they should be let go, but he was bearing significant cost, if he let them go, maybe there might be backlash against him, a new manager has no such problems and he can easily let these employees go. And so you would imagine intuitively that this sensitivity to separations as a function of disclosures would be more severe excellently for acquirers that are more sensitive to this disclosures. For example, if Christie were an acquirer, and it turns out that, before she acquired The Emmanuel Firm, she has a habit of letting go of employees that have these disclosures, because she knows that it's costly, it is easy to imagine that after she acquires Emmanuel, she's gonna implement the same thing. And so, what we're gonna gonna do is we're gonna try to get a sense of how strict Christie is before the merger, and to do so, we're gonna run these tests at the individual acquirer level.

0:31:14.4 EY: So we're gonna use the cross section of employees that work for a given acquirer, A and estimate this beta one coefficient, how sensitive is Christie in my simple example here [chuckle] to employee misconduct related disclosures. We're gonna sort acquirers and say, "Do those acquirers that seem to be more sensitive, are they the ones who are more for lack of a better word, trigger happy, more happy to let go of employees with this misconduct related disclosures after the merger?" And that's exactly what we find, right? Like HSD, there is like high separation for disclosures there that beta coefficient, when it's higher, those acquirers are more likely to separate with their employees. And we'll find evidence consistent with stricter disciplining mechanisms after the merger leading to this separations after the merger.

0:32:05.4 EY: Now, I laid out two possible mechanisms through which misconduct will drop following the merger. One was employee separations, the other if you remember were existing employees behaving better. And so, I've shown you results for separation here, and so how much do separations contribute to the drop that we showed earlier. We showed that it dropped from anywhere from 25%-34% depending on the merger, on the measure of misconduct. Is zero, for example, the right benchmark of the contribution of separation to that drop? Want to get a sense of just how much the separation contribute to that drop. We gonna do what I think is a pretty interesting exercise, so this... Then I come back to that. So what we're gonna do, is we're gonna say, look, let's look at the employees that either work for the target or acquiring firm and let's track them as before, right? What fraction of employees have a new instance of this misconduct related disclosure. Now we track the fraction from negative three before the merger to three years following the merger.

0:33:09.4 EY: Now, we're gonna do a counter factual exercise, we're gonna say, okay, now we know that you let Aaron go, we know that you fired him at T equal zero in the year of the merger. But let's assume you did not fire him, let's assume that you had kept him. If you had kept him and all the other employees that you had you let go, what your misconduct have looked like, right? Let's bring back. Those separated employees did not just vanish into thin air, a lot of them got jobs at other firms. So let's keep them back and assign their misconduct to the merged firm and see what misconduct would have looked like. And what we find is, if anything, you would have observed an increase in misconduct following the merger if you had kept the employees that you separated with.

0:33:51.8 EY: And so, this leads us to believe that most, if not all of the drop in misconduct following the merger is really through this channel of employee separations. Now, the thing that I skipped over before is really to show you that a lot of the separations are really happening for target firm employees, not necessarily for acquiring firm employees. Remember the earlier figure I showed you showed that most of the drop came in, in the first year after the merger. And So let's just zoom in in the months to the merger and look at the probability of separation here, separately for acquiring firm employees and for target firm employees. You can see for acquiring firm employees, there's almost no change after the merger, but for target firm employees, you can see a huge spike in the probability of separation especially for employees that have this misconduct related disclosures.

0:34:44.2 EY: And so, we conclude that this corporate control transactions can discipline the labor force. Employee misconduct declined by anywhere from 25%-34% following the merger and it is driven by separations especially by acquiring firm employees that are sensitive to these disclosures. Employees at the target have better misconduct record, however the sensitivity to separation increases following the merger, again, consistent with improved disciplinary mechanisms. Now, contrary to the market discipline hypothesis, the evidence really it is more consistent with the compliments hypotheses of like buying like, right? And it is interesting to know that M&A can be disciplinary even in the world in which like is buying like. Remember I said there was no specific prediction about the compliments hypotheses on employee behavior following corporate control transactions. And so even though the compliments hypothesis describes that evidence of matching and entry in to the market for mergers, it is nice to know that M&A can improve behavior, even in the world in which like buys like. Thank you very much.

0:36:00.9 CB: Terrific. Who wants to kick us off with questions? I can see you're talking, Professor Pritchard but I can't hear you.

0:36:18.6 Professor Pritchard: I was trying to figure out how to raise my hand. I've succeeded now, we're a year and a half into Zoom and I know how to work it. So my background with regard to this is from the broker dealer industry rather than investment advisor industry, but I think that there are a lot of similarities between the two fields. And when you were explaining the market discipline hypothesis, it's not your hypothesis it's out there, but it didn't make any sense to me because I understood from the broker dealer in the industry that the way you deal with employee misconduct is through firing, right? And there are differences among firms in their willingness to fire, overall I would say big firms are more willing to fire than small firms, and it doesn't make a lot of sense to acquire a firm and improve it by firing the advisors because you're buying a firm and then loosing a bunch of assets under management 'cause a lot of those accounts will leave when you fire them.

0:37:40.9 PP: So I wonder how much of the affect you have from mergers creating greater sensitivity. If a firm grew through organic growth, to a larger size and had a more intrusive compliance program, because larger firms have more rigorous compliance controls, would you see a similar increase in sensitivity to employee misconduct, right? Because if my intuition is the way you deter employee misconduct is by firing, and you vary on that basis based on whether or not you've detected it, and bigger firms can afford better internal controls. I'm not sure what my question is, that's my point.

0:38:43.4 EY: Yeah, it was more of a point. I was like, I didn't get a question out of that. Yeah. [chuckle]

0:38:48.2 PP: So well, is there a way of comparing your merged firms to firms that grew organically? 

0:38:58.9 EY: So one thing that we do in the paper that I did not present here today is we do a matching exercise when we're looking at the drop in employee misconduct, following the merger. So what we do is we say, let's take another firm say an investment advisory firm that was not in the market for M&A, and let's look at a firm that was in the same state, maybe had a similar level of this employee misconduct. And we match on three other variables. Now, let's do things the same thing for the acquirer. So we're creating essentially a pseudo merger here. Let's find an advisory firm that was not in the market at all, but that looks very similar to the acquirer. Now that we have this, our pseudo merge firms let's track their misconduct history alongside our merge firm, and what we find is that there is really no change in the pseudo merge in misconduct, in the pseudo merge merger pair. And the change really seems to come from the actual merge pair, if that makes any sense. So we do this matching exercise where we try to find, you know, a similar potential target, a similar potential acquirer. That should be just as large as the combined merge firm. And we don't find any change in disclosures there.

0:40:12.4 PP: But it's not as large. So they haven't had the change in internal controls, is my point.

0:40:20.0 EY: Oh, in this counter factual exercise...

0:40:22.9 PP: Your pseudo merged firm is not actually merged. There were no changes in the compliance department.

0:40:28.2 EY: Right. And the changes in the compliance department, that's our main point, right. Like when you have this two merged firms and there's changes in compliance, then suddenly you have this drop in employee misconduct, and against the backdrop of previous literature. If you read the prior literature you would get the sense that disclosures really go unpunished. Like the Egan paper, for example, says, half of advisors just get jobs with other firms, right. And so it doesn't... You don't get the sense that there is some disciplinary mechanism through which this disclosures get enforced. Even if you fire them, they can just find a job someplace else. And so, when you look at this corporate control transactions, we come in and say, look, this corporate control transactions really impose this discipline in a big way, rather than just like firing of employees at different firms.

0:41:15.0 CB: Professor Herman.

0:41:18.7 Professor Herman: Hi, it was a very interesting presentation. So just speaking back on the internal control improvements, one thing you might want to explore is internal whistleblowing or the information that sometimes the employees they pass it to Osher or other places that there has been financial misconduct, or there has been some other things. So those are some reported documentation that could also lead to some litigation. So maybe one of the channels that you were talking about, the separation whether that leads to an improvement in compliance or not, you can use this internal whistleblowing as a complimentary data for this trend in the argument. Maybe if that's any help, does any help to your channel that you might want to explore.

0:42:23.3 EY: Thank you, Abu. That's not something that we have thought about. We'll certainly look into that. I made a note of that.

0:42:30.0 PH: Thank you.

0:42:32.1 CB: Professor Yimfor. This is probably a dumb question, but my job is to make everybody else feel better about how much they know. So I don't... I'm not familiar with M&A generally. But I don't know... I'm curious, like when I assume that there are dozens of factors that acquirers are looking at when they're looking to acquire somebody and that this employee misconduct is just one small factor out of dozens. And so, I'm just wondering like how big is this? If you could provide me... Yeah. Like how big of a factor is this 5%-7% sounds like it could be a lot when you're talking about millions or billions of dollars, but I don't know if there are other bigger factors that... I don't have a sense of how big that is as compared to the other factors that acquirers might be considering.

0:43:34.8 EY: That's actually an excellent question. So in the paper we... To motivate essentially our focus on misconduct, we do two things. The first one is talk about its potential value relevance. By showing that, you know, these are some variables that this advisory firms will care about, like failures, levels of assets and the management. But in generally more specifically, your question is about, hey, look, if I'm looking if Christie is looking to buy the Emanuel firm, what are some of the top things that she considers? And so we read an industry report that suggested that 30% of these transactions fail because the acquiring firm didn't think the culture would be compatible. Right. And so when you think about what the culture is, right, that is very related to this. 'Cause there might be some firms that might be more tolerant. Maybe that's just their culture. They think you should be aggressive at all costs, even if you incur some of these disclosures and it might be other firms that don't think that way. They think, you know, maybe it's a negative. It's not good publicity when we have so many of these disclosures. And so the culture, the firm culture, as a reason for merging that really, again, led us to believe that misconduct is really relevant, not just in M&A, but for value in general. Yeah.

0:44:53.9 CB: Thank you. Someone else wanna chime in. It's okay. I thought he explained it really well, too, so if you all feel like you got it, you know. Okay. Thank you everyone for coming. Can we have a moment to acknowledge Professor Yimfor for his talk today? Either with your virtual icons, or you can turn your camera and clap. Oh, I like it, the little party signs. Thank you so much for participating today and for joining us. Please come back next month. I do have one little promo, because we are in the business of promoting side projects which is I'm hiring. So most of the center has about 20 research assistants from five schools who work with us each semester. We lost a huge... More than half of our group last year to graduation. And so, I am still hiring two law students and I'm still hiring a Ross MBA or possibly a PhD student to work on some projects. The postings are listed there. If you know some students who would be fantastic, please have them hurry up and apply because I'm starting to get nervous. So thanks again for that. And I hope that you will join us again at a future talk and with that, we will return to all of you 11 minutes of your time. So thanks everyone.

0:46:33.7 CB: Professor Yimfor, if you don't mind to hold on a second, I wanted to ask you about something else.

0:46:38.1 EY: All right.

0:46:42.7 CB: Jeremy, will you stay too.

0:46:48.3 Jeremy: Emmanuel, I just wanna say thank you for being here and I apologize for not being there for most of the talk. I had a conflict earlier. And so I'm sorry to miss, you know, it's tough when you only jump in for Q&A. [laughter] I mean, you made the talk sound very interesting. It makes me interested in wanting to find out more about the project. So I'm sorry to have missed it, but...

0:47:07.1 EY: Oh, thank you very much. Thank you.

0:47:08.9 Jeremy: I don't think we've met in person, but I'm, you know, at Ross as well. I'm the co-faculty director of the center. So I'll look forward to having you involved in more center projects as we go forward.

0:47:22.5 EY: Alright.