Baruch Lev and Feng Gu on Data Driven Mergers and Why Most Deals Fail
Andy Leonard and Frank La Vigne are joined by experts Baruch Lev and Feng Gu to uncover the complexities and pitfalls of mergers and acquisitions. We’ll discuss the controversial “killer acquisitions” in the pharmaceutical industry, which regulators fear stifle innovation and harm public health.
Our guests will share insights from their upcoming book, “The M&A Failure Trap,” which critiques current acquisition strategies and introduces a unique 10-factor scorecard for assessing potential success. From data analysis on 40,000 mergers over 40 years to the challenges and market trends affecting merger outcomes, we’ll explore why up to 75% of mergers fail and how decision-makers often benefit at the expense of employees and shareholders.
Whether you’re an entrepreneur looking to navigate M&As or a data enthusiast curious about the numbers behind these strategic moves, this episode offers a data-driven look at the forces shaping mergers and their real-world impacts.
Show Notes
The M&A Trap Book Link (no affiliate) https://www.amazon.com/Failure-Trap-Mergers-Acquisitions-Succeed/dp/1394204760
Highlights
00:00 Exploring data science, AI, mergers with experts.
04:43 Extensive data-driven analysis of mergers’ failures.
09:22 Investment bankers pressure companies to finalize acquisitions.
11:15 Managers get bonuses for concluding acquisition deals.
14:26 Global economy affected; star performers leave.
17:32 Mergers often lead to employee departures, layoffs.
20:24 Managed data engineering team during Unisys acquisition.
26:28 Analogies highlight misapplication of causal thinking.
28:58 Complex model reveals hidden variable impact.
31:01 Correlation can mislead; avoid single-focus traps.
37:14 Comprehensive analysis of acquisitions and their impacts.
38:39 Analyzed LinkedIn data on employee turnover trends.
41:50 Creative metric developed for private acquisition premium.
46:01 Acquisitions are widespread, impacting various industries significantly.
52:11 Unique 10-factor acquisition scorecard predicts success.
55:45 Deep dive into mergers and acquisitions data.
Speaker Bios
Baruch Lev is a professor emeritus at NYU Stern School of Business, where he has taught and conducted research on mergers and acquisitions for decades. He worked formerly at UC Berkeley and the University of Chicago. His work has been widely cited in academic and professional circles (over 63,000 Google Scholar citations), and he is a leading authority on corporate finance and valuation.
Feng Gu is a professor of accounting at the University at Buffalo and has extensive experience in analyzing the financial aspects of corporate acquisitions. His research focuses on the economic consequences of corporate decisions and has been published in top-tier academic journals.
Transcript
Welcome back to Data Driven, the podcast where we explore the big
Speaker:ideas in data science, AI, and all things data
Speaker:engineering. Today, we're diving into the complex world of
Speaker:mergers and acquisitions where data meets corporate strategy
Speaker:and not always in the friendliest way. With us are 2 top tier
Speaker:experts who know this landscape inside and out, Baruch Lev,
Speaker:professor emeritus from NYU, and Phong Gu, professor of
Speaker:accounting at the University of Buffalo. We're going to unpack why
Speaker:up to 75% of mergers fail and how to spot the ones
Speaker:that succeed. Buckle up. It's data driven insight at its
Speaker:finest.
Speaker:Hello, and welcome to Data Driven, the podcast where we explore the
Speaker:emerging fields of data science, artificial intelligence, and, of course, data
Speaker:engineering. Today, we're gonna talk about a branch of, I
Speaker:guess, applied analytics, where we analyze how
Speaker:mergers and acquisition data, goes through. And with us, we
Speaker:have 2 esteemed guests today. It's not every day we have 2 guests.
Speaker:So I'm gonna read the bio of 1, and Andy will
Speaker:read the bio of the other guest. With us today is Baruch
Speaker:Lev, a professor emeritus at NYU Stern
Speaker:School of Business, where he taught and conducted research on mergers and
Speaker:acquisitions for decades. He worked formally at
Speaker:UC Berkeley and University of Chicago. His work has been widely cited
Speaker:in academic and professional circles,
Speaker:and with over 63,000 Google Google Scholar
Speaker:citations. He's a leading authority in corporate finance and valuation.
Speaker:And also with us is Feng Gu. He's a professor
Speaker:of accounting at the University of Buffalo and has extensive
Speaker:experience in analyzing the financial aspects of corporate
Speaker:acquisitions. His research focuses on the economic
Speaker:consequences of corporate decisions and has been
Speaker:published in top tier academic journals. Welcome
Speaker:gents. Thank you. Thank you for having
Speaker:us. Yeah. Thank you for the invitation.
Speaker:Yeah. No problem. We're we're always great to to to have you here. And
Speaker:and part of our listeners are wondering, hey, I thought this was a data science
Speaker:podcast. And and I would say that if you are
Speaker:having an IT career, not just a data career or any career, you are
Speaker:gonna be impacted at some point along, by a merger and or acquisition.
Speaker:Sure. And I don't have a lot I don't know about you, Andy.
Speaker:I don't have a lot of fond memories of them all working out. It's
Speaker:always been a change, and, you know, change change is always,
Speaker:brings challenges. Yes. And I'm sure these gentlemen,
Speaker:study those challenges and have a lot to share with our audience,
Speaker:and us. You work for a large company in
Speaker:IT. I own a small boutique consulting
Speaker:company that that provides data engineering and and and
Speaker:similar services. So I'm excited to learn what you got going
Speaker:on. In case someone wants to acquire, my
Speaker:company. And I'm sure you're keeping an eye on this, Frank,
Speaker:in case someone wants to merge with yours. Well
Speaker:and, again, I wanna be clear. The current company I work
Speaker:for, I joined post IBM acquisition. Right? So all of these horror
Speaker:stories are actually the worst merger I was ever privy
Speaker:to was, as an employee, was, well, I guess I can say
Speaker:it now, the Bankers Trust Deutsche Bank acquisition,
Speaker:which, Deutsche Bank being a German company,
Speaker:when they passed out and and Bankers Trust was an American company. When they passed
Speaker:out the cards announcing the merger or celebrating the merger, the
Speaker:I speak German, so the English sides called it a merger.
Speaker:The German side used the word Uber Nemen, which means
Speaker:takeover. That's yeah. I know just enough Latin to
Speaker:pick that up. Which was, which I thought was
Speaker:interesting because that's basically what it was. So to when I talk about my
Speaker:merger horror stories, I'm not talking about where I am now. This is 20 years
Speaker:back. And, the other thing as a
Speaker:customer, when the the the companies
Speaker:I use have merged, I've not really been a happy customer. I think Sirius
Speaker:XM, XM Radio was a much better
Speaker:satellite radio provider than Sirius XM is. And that's just my
Speaker:opinion. That's not the opinion of anyone else. My wife seems to
Speaker:enjoy it, but it is what it is. So what really excited me about this
Speaker:so before our listeners start, like, what the heck are we gonna talk about? These
Speaker:guys are gonna bring data to the table, and that's why I'm excited to have
Speaker:them there. So I'm gonna get off my soapbox because people don't wanna hear us
Speaker:banging on. They wanna hear you guys.
Speaker:So starting with the data side, we
Speaker:have probably the largest sample of
Speaker:mergers and acquisitions ever assembled.
Speaker:We have a sample of 40,000 mergers and
Speaker:acquisitions worldwide,
Speaker:spending over the last 40 years. And
Speaker:on this huge sample, we
Speaker:have developed a quite sophisticated
Speaker:statistical model, multivariate
Speaker:statistical model with 43 variables
Speaker:to identify statistically,
Speaker:the attributes,
Speaker:the factors that contribute to success
Speaker:and failure of companies.
Speaker:Excuse me, of mergers and acquisitions. So, basically,
Speaker:the entire work that we did, which is summarized in the book, of
Speaker:course, is heavily data
Speaker:driven. It's also supported by
Speaker:other study, which are always
Speaker:data, driven large sample studies
Speaker:of specific issues of mergers and acquisitions
Speaker:that, we didn't examine.
Speaker:So, we combine all of this
Speaker:to a set of of
Speaker:observations and recommendations of
Speaker:why 70 to 75%
Speaker:of all mergers fail
Speaker:fail to achieve sales growth, fail
Speaker:to achieve synergies in in cost
Speaker:of sales efficiencies, failed to maintain
Speaker:the share price of the buying,
Speaker:companies. It's an amazing number that
Speaker:surprises most most people who
Speaker:see it. That that is a large
Speaker:number, and I'm kinda shocked to learn that. I would have
Speaker:thought that, you know, it would have been on the positive side of
Speaker:that that 5050 mark that, that the
Speaker:mergers and acquisitions succeeded, and there were benefits enjoyed
Speaker:by all. But it sounds like what you're saying is no about 3
Speaker:quarters of those fail on some or, you know, some or maybe
Speaker:all, of those desired outcomes. Yeah. I'm
Speaker:actually not surprised. I had heard that statistic before and kind of based on
Speaker:based on my anecdotal kind of personal experience, I think that that sounds reasonable.
Speaker:But the question I have is if if it if the situation is so
Speaker:bad, a lot of questions, How do they how do these companies convince their
Speaker:respective boards to take the buyout? Is it just a,
Speaker:how did how do they pull that off?
Speaker:The way to an acquisition is
Speaker:usually a failure of the acquiring
Speaker:company. Sales slow
Speaker:down, earnings turn to
Speaker:losses, market share is lost,
Speaker:and everything gets excited,
Speaker:particularly investors who are, of course, losing money
Speaker:and influential investors who have a a
Speaker:big say on company. Directors
Speaker:are are looking, and the
Speaker:call gets out of we have to
Speaker:do something big. And, usually, the
Speaker:something big is a big acquisition.
Speaker:And that's how that's how that's the usual
Speaker:way of getting, to this.
Speaker:Managers, are optimists.
Speaker:Lots of psychological studies have shown that
Speaker:managers are much above average
Speaker:optimists. Some of them are overoptimists.
Speaker:They may be they may be aware that many
Speaker:most, m and a, fail, but they
Speaker:are convinced that they will make it.
Speaker:And they are convincing their board of directors and
Speaker:sometimes even shareholders to, support it.
Speaker:Yeah. So the persuasion and
Speaker:the pressure to acquire also come from
Speaker:frequently, investment bankers,
Speaker:financial analysts, and consultants. These people, of course,
Speaker:say, you know, have obtained financial benefits
Speaker:from, completed deals. They always pressure
Speaker:the acquiring company to by pointing out, hey. This is a good
Speaker:deal for you, and we can help you, you know, go through
Speaker:this and clear all the hurdles and everything will work
Speaker:out fine. And, so this is really the
Speaker:best decision for you to make. They're really
Speaker:play these consultants and investment bankers really play a very
Speaker:important role in convincing, both sides of the
Speaker:acquisition to complete the deal as soon as possible.
Speaker:Gotcha. That sounds like sorry. Go ahead. I
Speaker:just want to say in conclusion, you know,
Speaker:some, m and a proposals are being
Speaker:rejected. Not everything is accepted. Just
Speaker:recently, an Israeli company
Speaker:got, an acquisition
Speaker:proposal from no less than Google for
Speaker:$23,000,000,000. Goodness. After
Speaker:after consideration, they, they rejected it. So
Speaker:not everything is accepted. But
Speaker:many, many acquisition strongly
Speaker:supported by the CEO are indeed
Speaker:accepted. Well, it
Speaker:sounds like there's financial incentive, for the
Speaker:people around the process for the process to
Speaker:conclude? Because I imagine they don't get paid unless the
Speaker:acquisition goes through. Correct? Yes. And there are
Speaker:also there are also quite large, incentives
Speaker:for managers for concluding the deal.
Speaker:A recent study showed that,
Speaker:many managers get, acquisition bonuses
Speaker:between $5,15,000,000.
Speaker:Got it. And that's for concluding the deal, not
Speaker:for succeeding, but for just
Speaker:concluding the deal. Wow. And,
Speaker:we have we have in the book, we show statistics,
Speaker:which I've never seen anywhere else, that,
Speaker:serial acquirers,
Speaker:their tenure is 4 to 5 years
Speaker:longer than CEOs that
Speaker:don't acquire or acquire just few companies.
Speaker:My guess is that, directors are
Speaker:somehow satisfied with very active CEO
Speaker:who try to change the course of the company,
Speaker:let them acquire our companies, and then they give them,
Speaker:more time to to somehow
Speaker:somehow, complete the complete the deal and complete
Speaker:the integration. But I was,
Speaker:someone someone just recently asked me, what surprised you most? One
Speaker:of the things that surprised me most in researching the
Speaker:book was this 4, 5 year,
Speaker:10 year edge of serial
Speaker:acquirer CEOs, irrespective
Speaker:of the success of the mergers.
Speaker:Yeah. And this difference of CEO tenure by 4
Speaker:to 5 years is obtained after we have
Speaker:controlled for other contributors to CEO tenure, like
Speaker:corporate performance and other important factors. So in
Speaker:other words, our conclusion basically says with everything else equal,
Speaker:if you make a series of acquisitions, your
Speaker:CEO tenure is going to be extended by 4 to 5
Speaker:years on average, which is really a
Speaker:long, long extension. Acquisitions are almost,
Speaker:tenure insurers or CEOs.
Speaker:So it sounds like the, the,
Speaker:incentives are a little bit lopsided.
Speaker:Yeah. Definitely are from all sides. As Frank mentioned,
Speaker:the, commission hungry, investment bankers, and
Speaker:consultants benefit from the deal.
Speaker:CEOs benefit from, the deal. The only
Speaker:one who paid the price are the shareholders. And and
Speaker:many times, employees are being fired.
Speaker:Customers, suppliers,
Speaker:suffer. Mergers have an
Speaker:overall effect on the entire economy
Speaker:on the Which I think this. Yeah. Which I think, like, begs the question,
Speaker:like, if you play this out long enough,
Speaker:more people lose than win. And, like, what's the effect of this in
Speaker:the global economy? Because a lot of during times of
Speaker:uncertainty, a lot of the star performers leave because they're not sure
Speaker:what's gonna happen to them. Yeah. Right? Because usually
Speaker:usually, the the acquiring company tends to keep more of their people.
Speaker:What and and and I think that's probably a different game if, you know, if
Speaker:a if an £800 gorilla buys a small start up. I think that's one type
Speaker:of dynamic. But if you have kind of these 2 industry
Speaker:titans that buy each other, right, something more
Speaker:akin to Deutsche Bank and Bankers Trust, right,
Speaker:there's probably a lot of because they see each one of them sees
Speaker:sees each other well, one side sees itself as a peer and the other
Speaker:sees it as superior itself as superior. And that's gotta lead
Speaker:to all kinds of weird personal interdynamics.
Speaker:Yeah. You're perfectly right. I
Speaker:mean, acquisition of large
Speaker:targets relative to the size of the acquiring
Speaker:company are almost, a recipe for
Speaker:failure. We analyze in the book the examples,
Speaker:several years ago of Sprint acquiring
Speaker:Nextel. That's the 3rd and the
Speaker:5th, at the at the time. The 3rd and the 5th,
Speaker:wireless operators. This was they
Speaker:were about the same size. Sprint was a little
Speaker:larger. This was an unmitigated,
Speaker:disaster, the whole thing. They
Speaker:they completely failed in in,
Speaker:merging the employees.
Speaker:They even they even kept the separate headquarters of
Speaker:the 2 companies and the separate operating
Speaker:systems. Customers will ask, do you want to
Speaker:join the operating system of Nextel or
Speaker:Sprint? I mean, huge churn,
Speaker:huge desertion of customers,
Speaker:and then the whole thing, collapsed.
Speaker:Yeah. Acquisition of large
Speaker:targets are very, very difficult to
Speaker:integrate. And you indicated most of the reasons, with your
Speaker:example of Deutsche Bank. Right. Right. And I'm a former
Speaker:Nextel customer. Same. And I was not
Speaker:I think I think the Sprint acquisition could have been worse. But if that's your
Speaker:metric, it could have been worse as from a customer's point of view. Yeah.
Speaker:I I suppose based on the numbers you're telling me, it could have been worse.
Speaker:It sounds like a pretty good pretty soft pretty safe outcome.
Speaker:I'm doing the low bar symbol. If you're watching the video, you could see that.
Speaker:But I'm the bar is down here for could have been worse.
Speaker:Frank, you asked about how this type of deals may
Speaker:affect employees of the target versus the acquiring company.
Speaker:I I think it's a great question. In the research for this
Speaker:book, we spent a lot of time looking into how
Speaker:acquisition deals may affect, employees.
Speaker:And we did look at, the reaction from the target
Speaker:company's employees, and we find that as soon as the news of
Speaker:mergers acquisition comes out, a growing number
Speaker:of target company's employees decide to leave the company. And
Speaker:this happened even before the merger, gets
Speaker:completed. So they learn from their experience
Speaker:or maybe from your experience involved in this 2 large bank merger
Speaker:that the target employees always get, you
Speaker:know, relatively unfair share in the post
Speaker:acquisition termination, for the purpose of
Speaker:creating synergies, cost savings, and so on. So
Speaker:on average, mergers acquisitions have not been
Speaker:friendly to employees. We're documenting one chapter of our
Speaker:books, the loss of job positions on
Speaker:average is about 5 to 7%
Speaker:of the combined entities workforce, which is a
Speaker:significant number. Yeah. You know, it sounds a little low
Speaker:when you put it that way. 5 to 7% doesn't sound like a lot. But
Speaker:I can imagine, you know, in these, you know, in in Frank's
Speaker:Bank, acquisition scenario.
Speaker:Yeah. That's, you know, that's across thousands of
Speaker:employees. Yeah. That can be a large number of
Speaker:people. Well, there was also the rock stars. You know?
Speaker:Yeah. I don't know how it is now, but back then, you know, Wall Street
Speaker:was very aggressive about getting you know, they would basically go to the top
Speaker:trader at, let's say, BT Bankers Trust, and say, hey.
Speaker:We know you're feeling a bit uncertain now. Why don't you have a conversation with
Speaker:us? Right? And you can you you'll make more you'll make,
Speaker:like, 20% more or twice as much and bring anyone you want over
Speaker:to. Right? So the I suspect the numbers are actually higher,
Speaker:but the published numbers in terms of layoffs are probably 5 to
Speaker:7%. But I think the star performers,
Speaker:I think you kinda lose the star performers almost right away. Right?
Speaker:Yeah. Yeah. You're you're perfectly right. That that's what
Speaker:economists call moral hazard, which
Speaker:means the employees employees you lose. It's not
Speaker:just a matter of numbers. You lose the best
Speaker:employees, those with the best alternative
Speaker:outside, and you are left with those without
Speaker:any or very attractive alternatives. So the
Speaker:degradation of the work workforce is much
Speaker:more serious than just the numbers. Yeah. Yeah. You
Speaker:know, I have an experience like that too. I I just, for some reason,
Speaker:it escaped me earlier, but I was a manager
Speaker:at Unisys, and I was managing the, the data
Speaker:engineering team. We called it the ETL for extract,
Speaker:transform, and load, team. There were about 40 people who
Speaker:were a combination of full time workers and
Speaker:then an extended collection of subcontractors.
Speaker:And we went through a merger and I'll spill the beans on this one too
Speaker:with Molina Healthcare that was headquartered in,
Speaker:out in California. And
Speaker:we had some of that. In fact, my my boss who was a
Speaker:director, he was a fantastic example of this,
Speaker:definitely a high performer, published 5 books,
Speaker:a known entity in the data field, and just an excellent,
Speaker:leader in my opinion. In
Speaker:the ramp up to the merger,
Speaker:or it actually was an acquisition. In the ramp up to that, he
Speaker:when he got wind of it, he began putting out feelers,
Speaker:for, you know, making a move to another company. And eventually,
Speaker:he did. And this was excuse me. His move, him
Speaker:leaving was a huge hit to the company, a
Speaker:huge loss. And he did this months before the deal
Speaker:was concluded, like a full quarter ahead of time. And does that I'm
Speaker:curious. Does that count in the 5 to 7%? Would his
Speaker:leaving count in that, or would you would it be post acquisition?
Speaker:In in some cases, it it is included. In other
Speaker:cases, it it may not be included. It all depends on the
Speaker:relative timing of acquisition announcement versus k. The
Speaker:fiscal year end. Because as you probably know,
Speaker:companies don't disclose the number of employees all the time. I
Speaker:think right now, they, you know, provide this number once a year in
Speaker:their annual report. So there's always some discrepancy,
Speaker:in the number of in the exact number of employees, you know,
Speaker:between fiscal year end and, the announcement of the
Speaker:acquisition. Gotcha. But on average, it should be, you know,
Speaker:around that number. You mentioned the importance of
Speaker:losing key talent. Frank also made the key point here. We
Speaker:completely agree with you. Actually, in one of the chapters in your
Speaker:book, we have a graph showing, clear
Speaker:evidence, supporting this effect of
Speaker:losing talent. We document that after the acquisition is
Speaker:completed, 2 to 3 years down the road, there is a
Speaker:clear pattern of declining employee productivity.
Speaker:So that's normally a sign of losing key talent.
Speaker:You know, you know, you have lost the most important human capital
Speaker:component of your combined workforce, and there's no way,
Speaker:your workforce productivity is gonna be as strong as they used to
Speaker:be. So that's clearly a consequence,
Speaker:on the on the employee side after mergers, acquisitions are
Speaker:completed. So I wanna mention we're recording this on the 18th
Speaker:October 2024, and the book is
Speaker:named the m and a, m ampersand a,
Speaker:failure trap. And the subtitle is why most
Speaker:mergers and acquisitions fail and how the few succeed.
Speaker:And that book is due out according to Amazon today.
Speaker:They're projecting November 15th. So a little less than a
Speaker:month from now is when that book is due to be available. Is that accurate
Speaker:as far as you know? Yes. Excellent.
Speaker:Now I'm gonna buy the book. So I wanna know more.
Speaker:Thank you. Thank you, William. Yes. We have one say
Speaker:say 1. Yes. We
Speaker:made it. Make it 2.
Speaker:Your order is going to be the most special one because it's the first one.
Speaker:And, and since you bought the book, you can all, you can
Speaker:also give us, high recommendation.
Speaker:Okay. As for And we'll do that. Yeah. Yeah. Well, both Frank and I, you
Speaker:may not know this, but Frank and I are published. We've written I Frank, you've
Speaker:written a couple. Right? Couple 3?
Speaker:3. 3. Yeah. Mhmm. And I've been involved
Speaker:either as the sole author or a member of a team for 14.
Speaker:But I started way before Frank to be fair. That's a great
Speaker:number. Well, it warms my
Speaker:heart to hear smart people say that, but I have to share. I
Speaker:have to share that it has way more to do with insomnia than
Speaker:intelligence. Just just so you know.
Speaker:That's even more incredible.
Speaker:I I recall holding, my my youngest is
Speaker:17 years old now. But when he was a baby, I did
Speaker:that year. I wrote 2 at the same time. I just wrote chapters in a
Speaker:book on a team, just a few chapters, but I'll never do that again.
Speaker:And I haven't since. But I was holding him and had, you know,
Speaker:my arm had his head in my arm here and holding the bottle, feeding
Speaker:him at, like, 2 AM. And I'm typing on the laptop with
Speaker:my other hand. True story.
Speaker:That's quite a story. Yeah. This looks like an an amazing
Speaker:book. I've yeah. I'm a data, you know, a data weenie, being a
Speaker:data engineer, and I've worked around financial data of all my career.
Speaker:What we did at Unisys was Medicaid, driven data.
Speaker:And so you get a lot of finance in there. So we get it you
Speaker:know, we dabbled in that part of it, and there's just so much financial
Speaker:data out there. And I've seen so many ways to analyze it
Speaker:and then ways to, you know, not intentionally,
Speaker:but misanalyze it. You you look at the data,
Speaker:an old story intentionally and intentionally. Well, I imagine there's some
Speaker:intent. I was trying to be nice, Frank. But
Speaker:I have an old story that I share with data engineers. It's
Speaker:not, you know, it's not a real life story, but it's an analogy of
Speaker:the misapplication of thinking that sometimes goes along
Speaker:with this. It's kind of a, you know, getting the cart before the horse or
Speaker:miss you know, misunderstanding cause and effect. And
Speaker:the analogy that I use is, if you analyze
Speaker:the altitudes of aircraft in flight, you
Speaker:will find that the altitudes drop as they near
Speaker:an output sorry, an output, an airport, and everybody says,
Speaker:well, duh. And I'm like, so one conclusion
Speaker:you could draw from that is in placing airports, someone
Speaker:did an analysis of this data and said, where the craft are
Speaker:lowest, we'll build an airport there. And we all know that's not
Speaker:true. You know? But Yeah. Yeah. That happens. That kind of thinking
Speaker:happens a lot in analysis. And I'm wondering if that
Speaker:kind of mistaken analysis, if mistaken cause and effect
Speaker:plays into some of the thinking early on. Is
Speaker:that any of that leading to the 75%
Speaker:failure or failure to achieve result rate?
Speaker:There are lots of studies that are done by particularly done by,
Speaker:consultants, and they are based on,
Speaker:simple correlations. For example,
Speaker:companies, high on the ranking of,
Speaker:ESG, made it through the COVID
Speaker:disaster better than, others. Gotcha.
Speaker:I, with a group of, other researchers,
Speaker:rather than looking at just the correlation between
Speaker:ESG and success, we used a big
Speaker:model that looked at, that had
Speaker:representation of the industry, other
Speaker:variables there. Turns out that,
Speaker:most of these high up on the ranking of,
Speaker:ESG, were high-tech companies.
Speaker:They were extreme they were extremely successful as we
Speaker:know, many of them. Yeah. And this,
Speaker:of course, was reflected in share prices and profits
Speaker:and others. And they also had the means
Speaker:to contribute to the community and do other things
Speaker:that those who rank companies on on ESG
Speaker:like. So this is a this is a
Speaker:clear example in statistics of the
Speaker:missing correlated variable. The variable
Speaker:that that really went in was the
Speaker:industry of, of, this. And and and these
Speaker:these, people who just ran the simple correlation,
Speaker:missed it. That's why we built we built a
Speaker:humongous model of 43 variables
Speaker:that attempts to take everything into account.
Speaker:And then when when one variable
Speaker:indicates success or failure, for example, in your
Speaker:case of Deutsche Bank, we have a variable
Speaker:of foreign acquisition. This variable
Speaker:comes out after the estimation with a negative
Speaker:coefficient, meaning it detracts. All the
Speaker:all other things equal, it detracts from the acquisition,
Speaker:success. So we can say with with,
Speaker:fair certainty that,
Speaker:this is indeed a contributing factor because we accounted
Speaker:for, for most of the others. Yeah.
Speaker:Yeah. Brooke is absolutely right about, the special care we
Speaker:take to ensure that we're not just documenting simple
Speaker:correlation. We're actually, you know, the identifying
Speaker:the cause and effect relationship, In most
Speaker:of all performance related variables, we
Speaker:make very careful adjustment for industry average
Speaker:performance, at the same time. So this removes a
Speaker:lot of confounding factors from our analysis and gives
Speaker:us a lot of confidence in the validity of our results.
Speaker:That makes perfect sense. And I can see, and
Speaker:you've got the word trap in the title of your book. I can see the
Speaker:trap of, you know, making
Speaker:a correlation, which is a valid thing. It's a valid point in my example
Speaker:about the planes and the airports. It's a valid example.
Speaker:Apparently, you know, what you're sharing with me is you're seeing this, and somebody just
Speaker:picking up and focusing on a single correlation
Speaker:and making that the driving metric. And
Speaker:that that makes perfect sense. And I as you were explaining that,
Speaker:both of you, I thought of, books I've read
Speaker:about Warren Buffett's, and his partner, and I can't
Speaker:nobody remembers his well, it's Charlie. Charlie Munger. Charlie Munger. Right.
Speaker:Him and Charlie work together, and they look at the fundamentals. And they
Speaker:just over and over again, they just pour through probably all
Speaker:of the things that y'all are recommending, you know, for
Speaker:people who are interested in a merger or an acquisition. You probably recommended
Speaker:the same stuff. It's, you know, the fundamentals of
Speaker:what makes a business, you know, stable. And as you
Speaker:mentioned, Baruch, about, Deutsche Bank, that
Speaker:foreign acquisitions number, that's not something I would have thought of. But,
Speaker:you know, if it's stored in a data table somewhere, then I'd I'd look at
Speaker:the data, of course. Mhmm. But it's not I'm not a business mind.
Speaker:I am a I'm an engineer, for better or worse. As someone who
Speaker:lived through it, like, there definitely was a lot of disconnect between American business
Speaker:culture and German business culture. Like, it was a very That makes sense. It was
Speaker:I mean, it was a massive disconnect. You know? Yeah. The joke we had at
Speaker:the time, I think Chrysler was bought by Mercedes or Daimler Group that year.
Speaker:Daimler. Around that same time. And the joke was, thank God that that happened
Speaker:because we would be the biggest cross Atlantic disaster.
Speaker:You know, everybody was so focused on we were a distant
Speaker:second compared to what's going on there. And that, I mean, if you Chrysler's never
Speaker:really recovered from that. Well, the the joke I heard about that
Speaker:is, you know, how do they pronounce Daimler Chrysler in Germany?
Speaker:And it was they call it Daimler. Yeah. It's slightly Chrysler is slightly
Speaker:yeah. It's true, though. Like and, you know, one card says take over, and the
Speaker:other side of the card in English says merger. Right? Like, it's it's it's,
Speaker:you know, a lot of people had a good laugh at
Speaker:that, but I mean, there was a lot of truth to that. And also too,
Speaker:like, there's a funny meme going around about this, where it was a
Speaker:professor basically saying a 100% of the people who don't understand
Speaker:the difference between causation and correlation will die.
Speaker:That's a good meme. Yes. I'll have to dig it up and and reshare
Speaker:it. This was this was,
Speaker:many, many years ago, and I took it to University of
Speaker:Chicago, a statistics course. One of the
Speaker:first example in the first class was,
Speaker:the instructor showing a very high correlation between
Speaker:lung cancer and living in,
Speaker:Arizona. No way. Of course of
Speaker:course, the correlation is there, but that's not the causation.
Speaker:Arizona's weather is very good for the lungs. And that's
Speaker:why lung patients are going to a
Speaker:result. Oh. So, the causation is
Speaker:exactly the opposite direction than what the
Speaker:correlation seems to show. Yeah. His his next
Speaker:example is that more people die in hospitals than at
Speaker:home, which means that which means that hospitals
Speaker:are extremely dangerous to people. I have to try to
Speaker:avoid try to avoid them. That's those are
Speaker:really good examples. And I I one of the examples I read a
Speaker:long time ago, I was gonna say it was from the it may have been
Speaker:from World War 2, but I'm not a 100% positive of that.
Speaker:But there were aircraft engaged in combat,
Speaker:and they wanted to reinforce aircraft to make them survive, you
Speaker:know, the engagements better. And since they were
Speaker:pointing out, the bullet holes are showing up in these patterns, and they
Speaker:noticed that, you know, there's some here and there's some that we need to reinforce
Speaker:those areas. And someone thankfully pointed out that, wait, these planes
Speaker:are making it back. We need to put the reinforcement where
Speaker:the where the bullet holes are not. You know? So
Speaker:yeah. Survivor bias. Right? I think that's That's yeah. Yeah. That's
Speaker:it. That's true. But, yeah, great examples.
Speaker:So you have to be careful with analyzing data,
Speaker:particularly in our case, and that's
Speaker:straight, into the topic of your,
Speaker:of your, podcast. Mhmm.
Speaker:I let I let, Feng briefly
Speaker:describe the many databases sources
Speaker:that we use and converge,
Speaker:to get this kind of a sample and statistical model.
Speaker:Yeah. Yeah. So this is, really, the most
Speaker:important part about how we did our research to write this
Speaker:book. Everything, as Brooke mentioned earlier, is data driven.
Speaker:Our main conclusions are supported by, you know,
Speaker:analysis using large sample, not just a couple of,
Speaker:case studies, some anecdotal evidence. No. To reach
Speaker:that level, we pull data
Speaker:from a large number of sources starting
Speaker:from a mainstream mergers acquisition database,
Speaker:which gives a lot of details about both the acquiring company and
Speaker:a target company, the time of the announcement,
Speaker:the terms of the deal, and other interesting
Speaker:details like exactly what the the acquiring company CEO
Speaker:said about, his or her expectations
Speaker:for the forthcoming acquisition and so on. So we
Speaker:use that as the starting point to,
Speaker:collect as much data as needed. As Brooke mentioned,
Speaker:you know, we try to avoid simple correlation kind
Speaker:of scenario. So, in addition to industry,
Speaker:level adjustment, we also look at entire
Speaker:history of the acquiring company and the target company, you know,
Speaker:3 to 5 years before they get to the point of making a
Speaker:deal. Try to understand the circumstances of the acquisition.
Speaker:And then that is completed by
Speaker:using financial statement data, which is obtained
Speaker:from the company's financial statements, across multiple
Speaker:years, both before the acquisition and after the acquisition.
Speaker:Of course, stock price, information plays a huge role in
Speaker:understanding, both investors' immediate
Speaker:reaction to the acquisition news, and the performance
Speaker:of the combined entity after the acquisition is
Speaker:completed over several years down the road. Not just a couple of
Speaker:months, not just 1 year. We actually track, 3 to 4
Speaker:years after the acquisition is completed in
Speaker:order to obtain, a more robust and a
Speaker:consistent view of how the value of the company has been
Speaker:affected by the acquisition, is that value creation or
Speaker:value destruction? Alright. I also mentioned earlier
Speaker:about, you know, employee turnover. You asked you
Speaker:made a lot of good points about how mergers acquisition may
Speaker:affect, employees, not just everyday employee, but also
Speaker:key talent, of each organization. So
Speaker:we obtained very detailed employee turnover data
Speaker:from a database that is, I think, based
Speaker:on LinkedIn, information. So the original source is
Speaker:LinkedIn, which is probably, the most
Speaker:comprehensive database nowadays on employee
Speaker:turnover, very detailed real time employee turnover, not
Speaker:just, you know, once a quarter, once a year kind of information.
Speaker:So, we had very detailed,
Speaker:you know, in details a very detailed data
Speaker:on the trend of employee turnover. We look at it month by
Speaker:month to see exactly, how employees
Speaker:decide to stay or leave, once
Speaker:the merger news, comes out. So that gives
Speaker:you a snapshot of, the variety of databases we
Speaker:use, to, you know, conduct our analysis
Speaker:and then to provide our evidence. It's it's really a very,
Speaker:very comprehensive process. But you mentioned
Speaker:LinkedIn, and, I'm pretty sure the grain
Speaker:of their, to and from dates of employment, That
Speaker:that is a monthly drain that that they store that data in. That's
Speaker:something a data engineer would pick up on. But I I
Speaker:love the way you're describing how you acquired your data
Speaker:and, you know, in that it was a very
Speaker:macro process. You were looking at as many companies as you could
Speaker:find. I like that part of it. I like the time span that
Speaker:you applied going 3 to 4 years after the merger acquisition
Speaker:occurred. It it really reminds me I mean, I'm more excited about
Speaker:reading the book now because it reminds me of the business books that
Speaker:I learned the most from. And I I won't mention the other books,
Speaker:but there's only a handful of them that take that approach.
Speaker:And I I think it bodes well for the success of your book.
Speaker:So I'm I'm curious how, if how
Speaker:and if you, encountered data
Speaker:that you either decided was out of bounds?
Speaker:Did you did you have limits on that? Did you run into
Speaker:any data quality issues?
Speaker:Yeah. In some cases, because we require the post
Speaker:acquisition performance information to be available for
Speaker:3 to 4 years after the acquisition. You know,
Speaker:some companies don't survive that long. Actually, we have seen
Speaker:cases where the acquiring company later on, became
Speaker:too weak and eventually being acquired by other company.
Speaker:So those cases were probably not fully captured.
Speaker:We also don't have full information on some of the
Speaker:private targets. We don't know everything about their
Speaker:performance, before the acquisition like sales,
Speaker:profitability, and so on. And, of course, these private targets
Speaker:don't even have stock price information. So you
Speaker:can't see how investors react, the investor of the
Speaker:target company reacts to the news of acquisition. You can't even
Speaker:measure, this frequently used metric called,
Speaker:acquisition premium. You know, in in case of, a publicly
Speaker:traded company acquiring another publicly traded company, you
Speaker:can easily measure this acquisition premium
Speaker:by comparing the stock price of the target before
Speaker:the acquisition use, with the deal,
Speaker:the the the acquisition price that the acquiring company decides to pay.
Speaker:But in the case of a private target, you really cannot do that
Speaker:because, you know, they don't have stock treated, on the open
Speaker:market. So we had to be creative. Brooke
Speaker:and I developed a measure relating the acquisition price
Speaker:to the sales number of the target, which
Speaker:is actually very useful information because this
Speaker:allows us to get around this private target issue and
Speaker:make the metric much more comparable. And we
Speaker:actually developed a lot of insights from using this, different
Speaker:measure of acquisition premium. Cool.
Speaker:That's interesting. That's interesting. I like the fact that you take a data
Speaker:driven approach to this. Right? Because you listen to Bloomberg or whatever, they always
Speaker:show the rah rah. Look how great this merger is
Speaker:gonna be. It makes sense in this point of view. And if you're lucky,
Speaker:maybe they'll spend 10 seconds on, like, the detractors of it and things like that.
Speaker:But, you know, looking at this data all up, like,
Speaker:it it seems that and also think, too, the other thing to
Speaker:double click on is, if it's a private company, it's probably
Speaker:going to be way smaller. So I think a bigger fish eating a smaller fish
Speaker:is less likely to have indigestion, so to speak.
Speaker:Whereas if 2 big fish eat each other,
Speaker:there there's a lot of territorial fighting.
Speaker:Yeah. That's that's exactly, what Brooke mentioned earlier.
Speaker:Acquisition of a larger target is much more difficult to succeed
Speaker:because the integration process can become very contentious.
Speaker:Fight of egos and, a lot of, you
Speaker:know, emotional issues can get into the way to
Speaker:prevent the integration to be fully successful.
Speaker:Right. That Right. That makes sense. And it it gives me hope as a,
Speaker:you know, as a smaller company that maybe one day someone will come
Speaker:along. And I keep up with a touch of newsletters on this, not
Speaker:not a lot. I really didn't start looking into it until we started approaching,
Speaker:the 10 year mark. And one of the things that
Speaker:shocked me was the size of of companies.
Speaker:And and when I talk about the size, I mean, how small
Speaker:companies are, revenue wise. I mean,
Speaker:I I saw one newsletter that was talking that a
Speaker:I don't know how big of a segment this is for targets of
Speaker:acquisition, but they were half a1000000 to a1000000 and a half in gross
Speaker:sales. And that was shocking to me. I was like, I would be thinking they
Speaker:were looking at 10, 20,000,000, you know, size companies.
Speaker:But according to this one newsletter, it
Speaker:was a hot thing, you know, going after companies that that size
Speaker:in revenue. And I was shocked. Can you
Speaker:still hear me? Yeah. I can still hear you. No problem. We you
Speaker:disappeared a little on the video, but Yeah. Because I I I got the
Speaker:phone call. No. I wonder. But if you if you can hear
Speaker:me, that's that's okay. Yeah. That's good. We can hear you. Strong.
Speaker:Yeah. Yeah. Yeah. So so speaking of small
Speaker:acquisitions, what you said is exactly chewing
Speaker:some specialty sectors. Like, in our book, we mentioned
Speaker:large pharmaceutical companies acquiring much, much
Speaker:smaller, biotech firms in order to beef up their
Speaker:product pipeline. You know, the smaller size of
Speaker:this target is really misleading, you know, when you mentioned
Speaker:sales because, these are basically start up companies
Speaker:and they focus on developing technology.
Speaker:Especially if you look at the earnings, many of them don't have profit for
Speaker:decades. But that doesn't mean they're not valuable. We
Speaker:actually have some cases showing that a large pharmaceutical
Speaker:companies are often willing to pay a very high premium to
Speaker:acquire these, startup biotech firms because they see the value
Speaker:there. So, you know, acquisitions coming all color
Speaker:and shades. It's it's a huge phenomenon no matter
Speaker:what type of industry you look at, not just in tech industries. If you
Speaker:look at some of the highly matured industry like food,
Speaker:energy, Every year, you see large and small
Speaker:deals all the time. So that's that's what really, you know,
Speaker:interest Brooke and I when we decide to,
Speaker:write a book on this topic because it's ubiquitous
Speaker:and affects everybody, not just shareholders, affects
Speaker:employees. In some cases, affects consumers,
Speaker:customers as well because, you know, a merged company
Speaker:may decide to increase price in order to show,
Speaker:the value of the acquisition. Right? Or decrease their
Speaker:services or downgrade their services. Move one
Speaker:of the levers on the seesaw there. Yeah. Yeah.
Speaker:Yeah. We have we have, on this point, we have
Speaker:a a brief chapter in the book, titled
Speaker:killer acquisitions. And these are the cases.
Speaker:Yeah. And we give examples. These are the cases in which
Speaker:the acquisition is made, basically,
Speaker:to kill the target in this case too. I've heard of
Speaker:that. Yeah. Yeah. The most the most probably the most
Speaker:the most famous case is Visa,
Speaker:trying to acquire Visa Visa debit,
Speaker:not Visa credit. Visa debit, which has
Speaker:a huge market share. I think they have 70% of,
Speaker:all the all the US market in this case. And here
Speaker:comes, a small start up, which
Speaker:is much more efficient in obtaining data,
Speaker:linking to customers and things like this. Mhmm.
Speaker:And they, they try to, they try to
Speaker:acquire this company, with the with
Speaker:the clear it was. It it came out in an email from the
Speaker:CEO with a clear intention to basically,
Speaker:terminate the, the product. The whole
Speaker:thing the whole thing was litigated by Department of
Speaker:Justice and then Visa retreated.
Speaker:But, we quote a study on the pharmaceutical
Speaker:industry, a very, very in-depth,
Speaker:study that, that
Speaker:track the products of the acquired company
Speaker:match with the products of the buying,
Speaker:company, they concluded about 70%
Speaker:of acquisition in the pharmaceutical industry,
Speaker:killer acquisitions. Because if you look after
Speaker:the acquisition, all of a sudden, you see that the product
Speaker:of the of the target disappears.
Speaker:And Gotcha. What are what are regulators' thoughts on that?
Speaker:Like, I imagine that Very, very negative.
Speaker:Very negative. In this case, of course, of pharmaceuticals, it
Speaker:affects health of people. Right. Yeah. It
Speaker:it harms it harms, innovation.
Speaker:And, this this is this is an interesting chapter.
Speaker:Killer acquisitions. I'm so looking forward on
Speaker:its own. Right? I'm I'm definitely looking forward to this.
Speaker:November 18th, you say? It's 15th. 15th. 15th, 15th. November
Speaker:15th, according to Amazon. Oh, no. Actually, now it just changed.
Speaker:I am not making this up. 13th is what I'm seeing now. It's nice. Oh,
Speaker:nice. Okay. 3 to 6. I maybe I misread it before. I thought you said
Speaker:13, but it says 13 now. That's the, the date given by the
Speaker:publisher. Yeah. So the the book is now being,
Speaker:I believe, produced in the last, phase of
Speaker:production. And then at the end of the month, we'll leave the warehouse. And
Speaker:around 13th November, it will be available for shipping. Yeah.
Speaker:Very cool. And then What I'll do what I'll do is I
Speaker:will put the link to the Amazon, page for
Speaker:the book. I'll put that on a calendar note and schedule it for, like, 5
Speaker:AM or something on 13th. And so I can go over and buy it right
Speaker:away. You you you really wanna be the first one to order. I
Speaker:well, I won't order it, but I'll buy it first once it's once it's released.
Speaker:I would I don't do the preorder so often because and what I'll do is
Speaker:I'll check probably on 10th to see if it's available by Kindle because I
Speaker:know sometimes they send those out a little earlier. Oh, yeah. That's true. And, yeah,
Speaker:I'll grab it then for sure. But, yeah, those If you're going to buy
Speaker:it, what are you going to do with the baby? No. That's true.
Speaker:That's a good thought. I don't I don't think the baby's a baby anymore.
Speaker:No. He's he's driving now. So
Speaker:that's a good good point, though, Baruch. Thanks. Thanks for reminding me. I need
Speaker:to stay on top of that sort of stuff, and I need all the help
Speaker:I could get. Yeah. We think we
Speaker:ran out of time for questions, but that's fine. I think this was an exciting
Speaker:conversation that I think explains a lot of what we're seeing in in
Speaker:our careers where we we start one company. You're also starting to see a
Speaker:pattern of, you know, let's say Microsoft buying LinkedIn. LinkedIn has
Speaker:largely been left alone. Yeah. Yeah. Yeah.
Speaker:Well, they were doing a lot right to start with. Right. Right. Right.
Speaker:Right. I think that's that's an interesting thing is that, you know,
Speaker:smart companies, they know if it's if it's big enough and it's doing the right
Speaker:thing on its own Mhmm. Leave them alone. Yeah. That's that's
Speaker:that's the story of Google and YouTube. Yeah.
Speaker:Yeah. Yeah. Yeah. Yeah. I'm here in the country. I I live out in
Speaker:the woods in Virginia, and we say if it ain't broke, don't fix it.
Speaker:Yeah. But on on the other hand May I may I
Speaker:mention one thing that, didn't come up with, in
Speaker:the discussion? We developed in this
Speaker:book and with about a large chapter to
Speaker:it something which I think is really unique,
Speaker:and that's a a 10 factor
Speaker:scorecard for acquisitions. Nice.
Speaker:Everyone knows that, lending decisions, credit
Speaker:card decisions, largely made by looking at
Speaker:at, the credit scores of people,
Speaker:we developed, based on the 10 most
Speaker:influential variables of our model,
Speaker:we developed a very easy to use,
Speaker:friendly to use scorecard that, you
Speaker:can you can before the acquisition, you
Speaker:can get a a a the likelihood
Speaker:of success of this acquisition, a percentage,
Speaker:which will indicate the likelihood of success.
Speaker:I guess that this would be very useful, both
Speaker:to managers in somehow early
Speaker:screening of several acquisition candidates
Speaker:and to, investors who are
Speaker:often asked to vote on acquisitions without
Speaker:any information. Mhmm. So this
Speaker:this, acquisition scope, is something
Speaker:which is, really unique to our book. Yeah.
Speaker:Yeah. And I would say as a entrepreneur who, you know, wouldn't
Speaker:mind somebody sweeping in and acquiring the company, this could
Speaker:help me improve my score. Yes. You know, make me a more
Speaker:attractive target for acquisition. Yep. You know, I'm not not saying
Speaker:any of my customers listen. I'm not selling. Yeah. But,
Speaker:well, we'll let you know if that happens. But the, but,
Speaker:yeah, I mean, it's a an I think all around, that's just great.
Speaker:And I look again, one more reason to look forward to the book coming out.
Speaker:Awesome. Cool. Well, this has been a great yeah. Yeah. This is
Speaker:great. I'm I'm really glad we got into this, and you've answered a lot of
Speaker:my questions about how acquisitions get
Speaker:approved, who wins, and who loses. Usually, it's the employees and the
Speaker:customers, and and who wins. And turns
Speaker:out that the people calling the shots are the winners. Funny how that works. I
Speaker:know it technically speaking, it's a correlation. But
Speaker:I see what you did there, Frank. You see what I did there? Where can
Speaker:people find out more about the book? Do you have a does the web the
Speaker:book have a website, or do you guys have LinkedIn
Speaker:or anything? Not yet. Maybe maybe
Speaker:maybe we should, create it. Okay. Yeah. Amazon
Speaker:Amazon gives, a short
Speaker:description of the book. Okay. Mhmm.
Speaker:And and the endorsement. We have some great endorsement,
Speaker:about this book. And, yeah. But may maybe the
Speaker:first, place to go is really Amazon and get
Speaker:the description of the book. Awesome. It is the
Speaker:m ampersandamanda failure trap.
Speaker:That's Got it. We'll make sure to put a link in the show notes.
Speaker:And anything else, Andy? No, sir.
Speaker:Alright. Well, with that, well, let's And that's a wrap for today's episode
Speaker:of data driven. We hope you enjoyed this deep dive into the
Speaker:data behind mergers and acquisitions, whether it's a friendly merger or
Speaker:an Uber name and take over. A huge thank you to our
Speaker:guests, Baruch Lev and Foam Gu, for their fascinating
Speaker:insights. If you've ever wondered why so many mergers
Speaker:fail, now you know data doesn't lie. Be sure to check out
Speaker:their upcoming book, the m and a failure trap, for even more
Speaker:data driven revelations. As always, thanks for tuning
Speaker:in. Don't forget to subscribe, leave a review, and
Speaker:join us next time for more data centric discussions.
Speaker:Cheers.
About the author, Frank
Frank La Vigne is a software engineer and UX geek who saw the light about Data Science at an internal Microsoft Data Science Summit in 2016. Now, he wants to share his passion for the Data Arts with the world.
He blogs regularly at FranksWorld.com and has a YouTube channel called Frank's World TV. (www.FranksWorld.TV). Frank has extensive experience in web and application development. He is also an expert in mobile and tablet engineering. You can find him on Twitter at @tableteer.