← The Brainy Business | Understanding the Psychology of Why People Buy | Behavioral Economics
The Brainy Business | Understanding the Psychology of Why People Buy | Behavioral Economics
477. The Financial Value of Gender Diversity in Business
00:52:55
In this episode of The Brainy Business podcast, Melina Palmer is joined by Dr. David Daniels, a Presidential Young Professor at NUS Business School, as they dive into groundbreaking research on the value of gender diversity in companies. David...
Loading summary
Transcript51 lines
- [00:01]
Melina Palmer
Have you ever wished you had more influence at work? That people would naturally be more likely to buy in on whatever idea you're selling them, whether they report to you or not? Well, you're in luck. I teach a virtual 10 week class on internal Communication and Change Management through Texas A and M University and it's enrolling now. Get details and enroll@hbl Tamu edu and click on Certificate program. You get to learn directly from me, including live virtual office hours over zoom with a cohort of interested brainy folks like you from around the world. Again, learn more and enroll in the internal communication and change management course at HBL TAMU. EDU. That's HBL like Human Behavior Lab, dot TAMU like Texas A&M University. EDU and click on Certificate program. Your future self will thank you and when you're ready, enjoy the show. Welcome to episode 477 of the Brainy Business Understanding the Psychology of why People Buy. In today's episode, I'm excited to introduce you to Dr. David Daniels. Ready? Let's get started.
- [01:25]
You are listening to the Brainy Business Podcast where we dig into the psychology of why people buy and help you incorporate behavioral economics into your business, making it more brain friendly. Now here's your host, Melina Palmer.
- [01:41]
Hello hello everyone. My name is Melina Palmer and I want to welcome you to the Brainy Business Podcast. Do investors value gender diversity in companies or does it not have an impact on where they choose to place their dollars? Is there a real bottom line value for companies who have more or less gender diversity or while this hasn't been able to be quantified or shown in a causal way for decades. New research from my guest today, Dr. David Daniels has an answer and I can't wait to share it with you. So who is he? David is a Presidential Young professor at NUS Business School. He holds a PhD in business administration and a Master's in Economics from Stanford. His work has been published in Organization Science, Proceedings of the National Academy of Sciences, Organization, Organizational Behavior and Human Decision Processes, Journal of Consumer Research and Research in Organizational Behavior and has been covered by media outlets such as Harvard Business Review, Time, Forbes, pbs, NPR, and NBC. His research focuses on influence negotiation, decision making, motivation as well as groups and organizational diversity. His primary approach involves developing and testing theories in natural field settings using cutting edge research designs like field quasi experiments and natural field experiments and large scale data sets to credibly estimate causal effects in natural real world contexts and thereby generating new insights which can advance theory and inform policy. He's won multiple awards including the Best Paper award from the Academy of Management, the Rising Star Early Career Award from the association for Psychological Science, and the Early Career Research Excellence Award from NUS Business School.
- [03:27]
Dr. David Daniels
School.
- [03:28]
Melina Palmer
Really quickly before we get into the conversation, I want to be sure you know that there are links in the show, notes for my top related past episodes and books, ways to get in touch with David and myself, and more. It's all within the app you're listening to and atthe brainy business.com 477 now let's jump right in. Dr. David Daniels, welcome to the Brainy Business Podcast.
- [03:52]
Dr. David Daniels
Hi, it's great to be here.
- [03:54]
Melina Palmer
Yay.
- [03:55]
Dr. David Daniels
I'm so excited to be chatting with you today. We had a nice little pre chat as always. And for everyone who doesn't yet know you. You know, before we jump into the research, can you tell just a little bit about yourself and the work that you do?
- [04:10]
Sure. So I got my PhD at Stanford's Graduate School of Business and now I am a business professor at the National University of Singapore's Business School in Singapore. And my research focuses on three major topics. 1 of them is diversity, which we'll be talking about today. And I also do work on motivation, especially pro social behavior, and then I also do work on influence and negotiation as well.
- [04:44]
Awesome. And we had a previous co author of yours, Polly Kang, on the show as well talking about the streak end effect. So we have a whole episode about that. But do you want to share a little bit about, you know, your draw to that research and that work?
- [05:03]
Sure. Talk about the streak end rule just a little bit.
- [05:06]
Melina Palmer
Yeah, sure.
- [05:07]
Dr. David Daniels
Okay. Of course, of course. So you should definitely check out the, the other podcast episode for the full details.
- [05:15]
Yeah, we'll give a little teaser here.
- [05:16]
Melina Palmer
Right.
- [05:17]
Dr. David Daniels
This is the.
- [05:17]
Okay, here's a little teaser. So there's this famous idea in economics called there's no free lunch. Right. And the idea is that if you're walking along the sidewalk and you see $100 bill on the ground, there's no point picking it up because it must be fake because if it were real, someone else would have already picked it up. Right. And so this is the idea that you can't get something without taking some loss on something else. And it's closely related to the ideas of optimization and Nash equilibrium and game theory. However, if you relax the assumption that people are sort of perfect, idealized, rational decision makers, you can find free lunches. And the streak end rule is a framework for thinking about free lunches. You can Find. So here's an example. If your day was going to have two hard tasks, so you're gonna. Your day would normally be, I'm gonna do one hard thing, and then I'm gonna do another hard thing, and then my day is over. That's gonna be a really hard day. You'll walk away thinking, wow, the day ended on a hard note, and I did a streak of two hard things. But what we show in our paper, called the streak end rule, is that if you add an easy task sandwiched between the two hard tasks, you'll actually feel much more motivated at the end of the day. You'll feel less burned out, more energized, and it's because your day still ends on a hard note. But you broke up the streak. So we went from hard hard to hard easy hard. And what happened is you ended up doing more work, but you feel better at the end of the day. Now that's a win win. And the same thing can be. The same idea can be applied to managers and employees. If you can design work in a wiser way than there are ways for more work to be done in a way that's better for the worker as well. And so I kind of want to leave it there, but one takeaway for now is break up hard streaks any way you can, even by adding more work, as long as it's easier. And that will be a win win for yourself and in our organizations as well. So I'll leave it there. Go check out the other podcast episodes.
- [07:58]
Love it, love it, love it. Yay. What a great explanation. And the free lunch is a different take on the conversation that I had with Polly. So that's super fun. Right. Get to have these different angles and perspectives on the research. And I just find that paper fascinating and it makes a lot of sense, and it's. It was really nice. Like you said of, you know, we can actually do more and feel better about it. Like, why. Why would you not?
- [08:29]
You can't. You. You can't. Why would you not? Well, the answer is it's not obvious how to do that right now.
- [08:35]
Melina Palmer
We.
- [08:36]
Dr. David Daniels
Yeah, right.
- [08:37]
It seems obvious when we're talking about it.
- [08:39]
Yeah, exactly. Right. But then, like, discovering those and for you. So actually, Polly and I do talk quite a bit about some ideas for that. So, yes, we'll check that out. We'll put a pin in it for now, and everyone will go check out the streak end effect after listening to this conversation, which is maybe not in an entirely different direction, but really talking about it's.
- [09:02]
Not entirely different. Actually, I, I should say one more sentence about the streak and rule. So I told you the takeaway, but I kind of buried the lead. So here's, here's, here's the one sentence. Here's the idea in a nutshell. You do a lot of different tasks as part of your job, and some of those tasks are easy and some are hard. But at the end of the day, you don't remember all the tasks you did equally. And the two sort of types of tasks that you never really forget when you're thinking about like a sequence of tasks are the final one, like the last thing I did today as part of my work, the final one, you think about it in a way that's sort of 50 times more salient than, let's say, the second to last thing you did. Okay. When you're evaluating your overall day. So that's the end effect. And then the other thing we show is we call it a streak effect. So if you got three hard tasks in a row, like hard hard, hard, or hard hard, hard, hard, which is a streak of four hard tasks in a row, the hard tasks in those streaks, if it's the longest streak you had, are also overweighted by a factor of about 50. So in other words, you may do many tasks in a day, but the ones that really stick in your mind after the day is done, when you're retrospectively evaluating the day, are the longest streaks of, let's say, hard tasks and the end task, the final one, everything else, you don't really think about it. It doesn't come to your mind. That doesn't mean you totally forget what happened. But that's not determines how you feel after the day is done. Okay, now I really will leave it there.
- [10:59]
I love it. I love it. And you know, for everybody, of course, the peak end rule is very closely related on this. And again, we won't dig in, but we have an episode on that as well. So let's talk about this new paper, this other paper you've been working on.
- [11:17]
Melina Palmer
I sort of hesitated to say new because I know you've been working on.
- [11:20]
Dr. David Daniels
This for a really, really, really long time.
- [11:23]
Melina Palmer
Yes.
- [11:24]
Dr. David Daniels
And you just finally got to publish and start to talk about it, which is exciting. So tell, tell us about the, the research and let's go back to the, to the beginning, I guess, or however you want to talk about it.
- [11:34]
Yeah. So there's a paper that is new. It came out relatively recently, but in some sense it's really old. And I Think the way you put it was perfect. We have to go back to the beginning. The genesis of this paper dates back to before I started my PhD program. So it was after I got in. I wasn't just randomly doing it, but it was before day one. And so. So I. So I was about to start, and I was thinking, okay, what was I interested in? So even back then, one of the three big things I was interested in was diversity. So I was. I was, like, reading papers. I was. And I was looking for, like, what are the animating questions that people really care about? And one of the most important questions in the diversity literature is the debate over what's called the business case for diversity. And so this is the question of does diversity cause better performance in our organizations? So, for example, if the gender diversity of an organization went up from like 30% women to 40% women, would that cause an improvement in performance? Would the company make more money? And would the workers become more creative, more innovative? Would decision making become better? And this has been one of the central questions in research on diversity for decades, for decades. And it seems a little hard to believe, but there's actually no clear answer yet. The debate continues. And the reason why is this so it has to do with causality. If we really want to pin down causality in the social sciences, the behavioral sciences, the medical sciences, we know how to do that, at least in principle. And the answer is you run a randomized experiment, right? So this is why we know that painkillers work and don't have crazy side effects that will also kill you. The idea is we take a bunch of people into our experiment and we randomly assign them to a treatment group or a control group. So we might give the treatment group Tylenol, and then we give the control group a placebo drug, which is a fake drug. And then if we have a large enough data set, we'll be able to detect that, yeah, Tylenol really does reduce pain compared to a fake control placebo drug. So that's how we learn about causality, mostly from randomization. The problem with diversity is you can't randomize it, at least not in major companies. Right. Google will never let me or anyone else randomize who works there. And that is an incredible barrier to learning about what diversity really does. And so that has been impossible. So what have people done? They've had to rely only on correlations, and correlation is not causation. So let me lay the groundwork for this discussion by giving a fact that there's no dispute about. There is a positive correlation between diversity, by which I mean demographic diversity, such as gender and ethnicity, and company performance. There's zero dispute about that. That's true, but what's not clear is why. So the business case for diversity hypothesis says that, well, one thing that's going on that contributes to that correlation is, yeah, diversity causes better performance. What that means is if we force companies to increase their diversity or if they, you know, if we cause them to do it right, then they would have better performance. It would cause them to have better performance. So that's one possibility, but there are other possibilities. So a second possibility is reverse causality. So it could be that better performing companies have more resources, more money, and that puts them in a better position to attract and retain more women and ethnic minorities. Now the second reverse causality possibility is definitely happening. There's little dispute about that. And it's, it's because we are confident that reverse causality is in the mix that makes testing the best case for diversity very hard. There's also a third possibility which is called omitted variable bias. So it could be that there's some third factor like how old a company is or what sector it's in that both causes diversity and causes performance. So financial companies, for example, they tend to earn a lot of money and they actually tend to be relatively gender diverse, at least in the junior roles. And so there could be omitted variable bias as well. So surprisingly, despite the fact this is like a first order question for every organization, for academics as well, people have want have been looking for answers for, for decades and decades, like probably like 40 years at least, the answer is we don't know. So I was reading this and on the one hand it's very depressing because a lot of smart people have really done their best and there are a lot of great ideas in the literature, but at the end of the day, when there are a lot of great ideas, like some great ideas are competing and they cannot both be true at the same time. And so we do need a causal test and that's what had eluded people. And so on the one hand, that's kind of depressing, but on the other hand, how you could, the glass half full side of it is I thought this could be an opportunity maybe if I could find a way to get at this question that no one else had done. And while I was thinking about that, something happened which is that Google released its first diversity report on May 28, 2014. So before May 28, 2014, no one knew who worked at Google. It was A secret. So everyone knew who was on Google's board because they have to report that in their filings. Right? But the overall workforce, no. So this is part of the reason there. We know a lot more about board diversity and what correlates with that, and we know so much less about workforce diversity. It's because board data has been circulating, you know, for forever, since a company goes public. But workforce diversity was a closely guarded secret. And it's not that no one cared. Actually, people really cared. So for example, like at some point when Google started to get big in the 2000s, CNN started knocking on their door and they wanted to talk about who works at Google. And Google said, oh, we're not going to tell you. And then they started knocking louder and they said, and eventually it came to a point where they said, okay, you have to tell us, and here's why. You're a federal contractor, so you accept federal money. And because you do that, you must abide by certain disclosure requirements. And among these requirements are you have to tell us who works here. And so under the Freedom and Information act, you legally must tell us what your workforce gender diversity is. That's what CNN said. And they filed a FOIA request and Google counterfiled and said, no, we don't. And there is a loophole that lets you wiggle out of a FOIA request if you're a company. And the loophole is this, you don't have to disclose information if it would be a trade secret. And because revealing a trade secret that's substantial could cause you significant competitive harm. And that is what Google said. It said that its workforce diversity numbers were, and I quote, trade secrets. And if it were to reveal them, Google would suffer a substantial, quote, competitive harm. And Google won. They won the legal battle there. And the same defense was used by other major companies that successfully blocked FOIA requests to get them to divulge their diversity numbers. And so a stalemate wasn't placed for years. Throughout the, like the late 2000s, early 2000s, until May 28, 2014, Google just did an about face. And no one knows exactly why. It's hard to know for sure. But my conjecture is that they saw the writing on the wall, thought that eventually these diversity reports were going to come out and they wanted to stake a claim and say, okay, let's try to get out in front of it and then we'll claim the mantle of leadership on this issue, or that's something we'll try to do. So what they revealed, when it came, when it came to workforce gender diversity was they were about 30 or 31% women. So Google's workforce around 30% women. And when they revealed that everyone slammed them. So they were, the numbers were slammed as like too low, below expectations. And specifically, and I think very importantly below the gender diversity of the talent pool, specifically meaning the percentage of people with, let's say, computer science degrees who were women, which was around 40%. So it was taken as a problem that 40% of computer science degree holders were women, but Google only had 30% women in its workforce, and that was specifically what they were slammed for. And so I was reading all this stuff and I thought it was super interesting and like, and I at some point I just like looked at what happened to Google stock price and it took a big nosedive and I thought, oh, that's interesting. And like, I sort of marinated with that for a few weeks. And then what happened after Google's first diversity report was it sort of forced the other big tech companies like eBay, LinkedIn and so on, Yahoo to start releasing their own first diversity reports as well. So those other companies that first diversity reports came out in the weeks and months to follow in 2014. And what I eventually realized was that I could build a database that sort of took the diversity numbers from every first diversity report and took the stock price reactions to every diversity report and lined them up. And my hope was that I could use that kind of data to back out the stock market valuation of workforce gender diversity and answer the question, do investors value workforce gender diversity? Do they think that firms have too little, meaning an increase would be good for the company or possibly too much, meaning that a decrease would be good for the company. And that is eventually what I ended up doing. So that's the genesis of the project. And so the genesis dates back to 2014. The project, the paper came out this year in 2024. So it took about 10 years.
- [23:26]
Melina Palmer
Yes.
- [23:27]
Dr. David Daniels
And I think it's, it's amazing. I love in this story knowing, like you say, it's the paying attention to the right thing at the right time. And you know, were ruminating on the question just, you know, right when something was announced and you were paying attention in a way that, you know, allowed you to see something that, you know, might have gone, been missed, you know, by others and even, you know, looking back now. So I would say one thing though too is for people, you know, older data, if you're looking back, there may be something of an announcement, you know, things you can learn from existing data sets if you take the time to be thoughtful about what that causation could be, right, to be able to, to look back, which is good, I think. Cool and inspiring to know that that exists. And you know, like you said, there's this really, you know, key piece of seeing what happens when that announcement is made. Let's talk a little about, like you talked a little on that expectation piece.
- [24:31]
Melina Palmer
Right.
- [24:31]
Dr. David Daniels
So, you know, we expected that you would have more than 30%. So we have a negative response. You know, is the, does the positive side see a positive lift for those organizations where you're expecting it to be low because everybody's at 30%, but we had 35 or whatever, you know, that happens to be, you know, how, how do those things go together? And of course, I know we'll get to this, but, you know, how does that translate into real dollars for an organization?
- [25:02]
Great. Yeah, you read my mind. So let me just lay out the main findings and then if that'll answer your second question, then I'll go to your first question. So, so here, here's the main punchline. So if you're a major S&P 500 tech company like Google or ebay, then according to my evidence, 1% more workforce gender diversity would get you about $152 million. So it's really like a big, a huge boost, like certainly bigger than I would have expected going to the project. And if you are an S&P 500 financial company like JP Morgan or BlackRock or something like that, then having 1% more gender diversity would get you about $18.7 million. So it's about seven times smaller, but still a very large number. So those are the headline findings. And I think that's another nice thing about using. There are several nice things about using stock market data. One is that you can tease out causality, as I suggested a little bit, and we might talk a little bit more about that as well. The other one is you can quantify the impact of like, how much is diversity worth? Not just is it valuable, like, does it have positive value or negative value in the eyes investors, but how valuable is it? Is a question that is surprisingly hard to answer in most research in like the social sciences. And so I think that one distinct feature of our research is that we can, we can quantify it. And I think that's nice because, you know, there are trade offs everywhere and even things that are very good, right? Like, like curbing global warming or, you know, any, any social good you can think of. Yeah, there are, there are diminishing returns. So yeah, so there, there's positive value for sure. It's not infinity. So what is it? Is. It is a question that I think is not asked enough. And so I think it's the strength of our research that we can answer it. So let me go back to your first question. Would you. Would you reset your first question? Just so.
- [27:18]
Sure. Right. Yeah. It was in the space of, you know, we saw some of that negative response. Right. So if we. Is the positive true and if so, you know, kind of how that comes together.
- [27:29]
Okay, great. So this is exactly the right question to ask. So I do a deeper dive into Google in the paper and ebay as well. So what ebay revealed it's first when it dropped its first diversity report. EBay's first diversity report revealed that it had about 42% women. So that was basically at parity with the labor supply in terms of gender diversity. And ebay got a stock price rise, so they got a pat on the back rather than a slap on the wrist. Now, ebay is also good, a good example of why this stuff can be tricky in practice. So if you go and look at the actual ebay stock price, when it released its first diversity report in July 2014, it actually goes down a little bit. So you may look at that and then email me an angry email and say, david, I looked and it's not true. And so what we do in the paper is we're not just using like the raw stock price data when you're using the methods that we use. So our method is called an event study or a financial event study. So what you have to do when you're testing for stock price reactions is you want to make sure that when you're testing the impact of an event, such as the release of the diversity report on a company's stock price, you want to make sure that the company is moving in a way that's not simply driven by the whole stock market going up or going down. So, for example, if you release your diversity report, like during the great depression, you know, that's not. It's going to look bad regardless of what happened in the diversity report. So it turns out for ebay that actually the whole stock market went down on that day and actually ebay went down less. So there's a positive stock price reaction that appears on the day of ebay's first diversity report. And so that's one of the subtleties of the research is that what you're looking for are what are called abnormal returns. So these are changes in stock prices that cannot be explained by broad movements in the overall stock market or in the company's industry, like all the tech companies. Right. So we're looking for divergence from the market. And actually we're looking for something even more specific. So for example, if you look at the day before Google's diversity report dropped, then, you know, you should kind of see its stock price more or less following the rest of the stock market. You know, more or less a random walk. And, and that is what you see. And the same thing for ebay. And what you want to see when you're looking for causality is things not only to appear, but appear exactly when they should and appear relative to something that's like a control group. So here we have a control group, which is the broader stock market. And so if you think about like tying, going from like over here to over here, we're looking for. So like, this is, let's say by top hand, this is Google stock price. The bottom hand is the S&P 500 or the entire stock market or all tech companies. So, you know, before the diversity report, there's like a random walk. There's zigging and zagging and then bam, let's see the diversity report hits here. And then Google stock price takes a plunge and the rest of the stock market keeps going along in the random walk. So you're looking for a divergence relative to other observations that act as a kind of control group. And the divergence should appear when the diversity report came out exactly on that day. Right. It shouldn't be like four days later because the market should price it in very fast or at least start to price it in very fast. Right. And the day before the diversity report, if the report wasn't leaked, you shouldn't see anything really. And that's exactly what happens for Google and for ebay. So you see a stock price reaction when you should, which is starting day zero, the day of their first diversity reports, it starts appearing on day zero. And then it keeps getting bigger, actually for about seven days. And it kind of then asymptotes out, which is actually a pretty slow stock market reaction. So there is a big reaction very fast, but there. But I think there's a lot of ambiguity about diversity. This is one of the hardest things for our market to price. Right. Like, let's compare it to like the death of a CEO who was really good. Now you know how to price that. You know, you know the stock price of a company is going to drop when that happens but for diversity I think it's not so clear. So what you see are you see the more risk loving investors, or at least risk neutral investors move first. And so that's where you're getting the initial like day zero stock price reaction. Almost everything in our paper is just about day zero, the immediate first day reaction. However, what you also want to see is that it doesn't just like blip for one day and then go immediately back or anything like that. And we don't see anything like that. It actually continues, it's called continuation or momentum and then it asymptotes out after about a week. So my interpretation of that, which is, which is somewhat speculative is that the market's really slow at pricing this in and but as it learns more about diversity it's going to get faster and faster anyway. So to wrap this up and Summarize your answer. EBay reveals that has relatively high, a relatively high number of women in its workforce and it gets a stock price boost. Google of course revealed that it had relatively low numbers women and it stock price drop. And it's not just Google and ebay. If you line all the tech companies up, it's pretty close to a line and the companies on one side get rewarded. The companies that reveal high diversity, relatively speaking and the companies on the other side get penalized. So this is not a case where we like only see penalties if it's bad, but then if it's high there's nothing. Right. It's not that we see both rewards and penalties, which means that investors expectations were somewhere in the middle, which makes perfect sense actually. Right, because if you're not sure, you're going to take a guess which is going to be somewhere in the average and then when information is revealed then you'll learn and your beliefs will go in one direction or the other. So that's exactly what should happen. That's useful to know both empirically because that's what happens and it also reinforces the interpretation of the results as it really is a reaction to the diversity numbers themselves. It's not, for example, just oh, I just don't like it when companies talk about diversity. So I'm going to penalize them all. It can't be because eBay was rewarded for with high diversity numbers. Google was punished for low diversity numbers.
- [34:41]
Definitely. Thank you for the explanation and for showing how that has worked. And I know you talked a bit about some of those financial companies in addition to the tech ones and I just want to call attention a little bit to Knowing that you don't have to, we don't have to just wait around for, you know, an industry to be willing to release their numbers on their own in a diversity report. Right. Because the Financial Times was able to really put a, you know, collection of information together. Can you talk a little bit about how that data set is different from what came from those tech companies?
- [35:17]
Yeah, yeah. So thanks for bringing this up. So I, the story I told, which was the genesis of the paper, which is why I led with it, was about the tech companies. But I actually. So like I said, if you're an S&P 500 tech company, 1% more workforce gender diversity is valued at about $152 million. In my data. I have a completely separate analysis using financial companies like JP Morgan and Bank of America. And there I find the same direction and the magnitude there is about 1% gender diversity among those financial companies is worth about $18.7 million. And the question is about the source of their diversity reports. So unlike Google and eBay, J.P. morgan didn't release its own, like J.P. morgan Diversity Report, at least not at first. So here's what happened. The financial times emailed 50 of the world's leading, like biggest financial companies and what they said in their email was, hello, we've noticed that you haven't, you've revealed little or no gender diversity data and we would like to know, so please tell us your gender diversity numbers. And just so you know, if you don't tell us, we're gonna, we're gonna reveal that too. And, and they got a pretty high response rate. So 35 companies out of the 50 said, okay, here you go. And 10 of those 35 companies were, were based, headquartered in the U.S. like JP Morgan and Morgan Stanley bank of America. So we, we did a completely separate analysis on those companies as well. And we find again, a positive pattern. Like I said, 1% gender diversity is worth about $18.7 million. And so there. The diversity report was actually a mega report released by the Financial Times that contained what you could view as diversity reports from 35 companies, 10 of which were in the U.S. so those are the ones we examined for comparability. For comparability with, with Google and ebay and the other US based tech companies as well. So it came out in a different way. But ultimately you can do the same kind of thing. And I'll say one more thing. So there are pros and cons of different natural experiments that the world hands you. So the one strength of having the tech companies release Their diversity reports on different days is that if one day is a little wonky, it's only going to affect that one diversity report. Right. Not the others. The con is that the companies did voluntarily release their diversity reports. So there's some self selection there. And so for the financial companies it was the opposite. So the Financial Times chose the day on which the diversity reports were released. That's good because it means it wasn't under the control of the companies. The con is that they all release on the same day. So there's a greater risk of like something being odd about that day. Now I should say we were very careful to scour the news for any other kind of news that might have been released on the same day as the diversity reports. And we did find this in one case. So Yahoo had some, there was some merger announcement that was like vaguely related to Yahoo that was revealed it dropped on the same day Yahoo dropped its first diversity report. So we tossed it from the data. We don't use it in any of our analyses. But we also tried not tossing it. It doesn't really matter. But that is to say it is in principle possible that like you know, they tried to bury their diversity report under some other good or bad news. We were very careful to look for those cases. We found one possible case that happened and when in doubt you throw it out. And then I also did something similar for like because I talked a little bit about where sometimes called placebo tests. So like the day before the report happens you shouldn't see anything going on in the stock market really because there's nothing that happened. You should find a null effect. Right. And the tricky thing there is like what if the day before the diversity report then Apple says oh here's our new iPhone. So then that's not a valid control placebo day. So we also had to drop those. So I became an expert on this like three month window around the diversity reports and there's, there's. Yeah, so but it doesn't matter that much. But, but, but it's something that, that you absolutely should do to be confident. You know, it's another way to be confident that we're really picking up an effect of the diversity numbers themselves, themselves and, and not something else weird. You know, like in any, anything you, any conclusion that's really important. You want to triangulate on that as much as you can and just like test it like see like try a weakened part of the analysis and see if things fall apart. Right. So this is why we, we do it from Every angle we can think of. So we look for any other news that could have possibly affected our analysis. If we found anything, we dropped it. We looked for nothing happening before the diversity reports and that's what we find. We don't want to see an immediate reversal. Right, because then there's no long term effect and there is no immediate reversal for Google and ebay at least. And there are many other things we do as well. So let me take it back to the beginning. So I talked about sort of the albatross around the diversity literature's neck, which has been reverse causality and emitted variable bias. So here reverse causality can't happen. Why? Because performance here is stock price performance. So it can't be that Google releases, you know, Google stock has some stock price change and that instantly changes the diversity of Google. It can't be. So it sounds kind of funny when I say it like that, but that actually is a big problem because what is, what normally happens is people look at correlations between, let's say profit and diversity and there is a positive correlation. But we don't know if diversity helps to boost profit or if being a more profitable company helps the company to attract more diversity. We don't know for stock prices. We do know. We do know. So we can rule out reverse causality. And omitted variable bias mostly kicks in with the concern that other like unrelated news could have happened on the same day, like the iPhone 17 or something. So that's why it's important that we sour the news, throw out anything that might be contaminated. And so that's why we did that. And so that's why we can be confident in causality even though we don't have a randomized experiment. Although if we did have randomized experiment, this project would not have taken 10 years.
- [42:41]
Yes, pluses and minuses, right?
- [42:45]
There are pluses and minuses. So the burden of proof is very high. When you don't have randomization, it's. And rightly so. Rightly so. Now I, as someone who does this kind of research, you know, there is a reward as well, which is that it's pretty rare. It's hard to do. And I understand why. I understand why it's rare. If nothing else, it takes a long, long, long, long time. You know, all research takes a long time. This kind of research really takes an extra long time. Yeah. And now that being said, I think like part of it is because, you know, diversity is a, is an ever changing topic and like the debate around it looks totally different today than it did Five years ago. And five years ago it was totally different than it was in 1995. So I think part of the reason why the burden of proof was so high is because we're still learning. We're still learning like what it means to people and like different ways in which it could matter. And I will probably talk a little bit more about this. So in the 1990s, diversity was mostly a compliance exercise. And you, you basically did it so you didn't get sued. Now that's still there today. But the dis, but the, the discussion, the discourse is much broader. And so one big change is that in the last 10 to 20 years, the business case for diversity has mostly emphasized, not that it's a hedge against lawsuits, but it is a tool for increasing your creativity, increasing your ability to innovate, and improving your decision making. And that is simply not the discussion we were having in 19. So it's an evolving, dynamic topic, and rightly so. My hope is though, that there's some solid causal evidence in this paper from real companies and that's something that we haven't seen before. And I hope that will act as some kind of anchor because this is far from the last word. I actually think it's closer to the first. And so there are so many questions that I don't answer in this paper, but the paper does plenty. If you read it, you'll see what I mean. So for example, this paper is about workforce gender diversity. And in the financial company's diversity reports that were revealed, or you might say, except exposed by the Financial Times, the financial company's reports only talked about workforce gender diversity, not about workforce ethnic diversity. But in, in some of the other reports by the tech companies, they did talk about ethnic diversity. And so I have a follow up project I'm working on about that. So I bring that up just because that's a closely related topic, but it's not the same thing. And so I think this paper that came out in organization science called Do Investors Value Workforce Gender Diversity? It's closer to the first of its kind of study. Now that being said, it really does build on the shoulders of giants. And so it is deeply connected to everything that's come before, especially theoretically, but just empirically, it's doing a different thing. So for example, so what I've talked about in the paper is what we show in real world field data. And I've talked about stock price reactions, but we do more in the paper. So a natural question is, okay, so investors place positive value on workforce gender diversity. Why, like, why do they think it's a good thing? Why do they think having more would be good? What do they think it does to an organization when its gender diversity goes up? And so what I wanted to do was not. I wanted to sort of unpack the psychology of investors and let them tell me why they were doing it. But of course, like, the researcher has to decide what questions to ask them. So my co authors and I thought about this for a long, long, long time. So I. Let me. Let me thank my co authors. So this is joint work with Jen Daniels, who's at Yale, Thomas Lees, who is at Northwestern, and Maggie Neal, who's at Stanford. So the four of us thought very carefully, you know, and because, like, at one point we're like, oh, like, should we just. Should we keep it focused and only ask about, like, okay, if a company says it has a lot of diversity, what do you think happens to its creativity? Do you think it's more creative or less creative? And so that seems obvious. Like, we have to look at that. But then, like, the. A lot of the classic stuff from the 90s talks about, like, being sued. So then I was like, okay, we have to look at that. And at some point we just decided that we have to look. We have to do a huge literature search, look at everything that's been done, and try to pare it down to, like a top six or something like that, A manageable number. And so that's what we did, which took several years, and we landed on a top six. And we ended up landing on three possible upsides of diversity and three possible downsides. So the three possible upsides we considered were, do investors think diversity is good for performance because it boosts creativity and innovation? Does diversity help you think outside of the box? Is it like the antidote to group think and conformity? Right. Does it improve decision making? That's one possibility. And we didn't want to take an empirical stance on that. We wanted investors to tell us what they thought. So we also looked at do investors think, okay, one thing diversity does is it reduces legal risks or, you know, negative scrutiny from politicians or regulators. Gets them off my back. So we looked at that as well. Another possible upside of diversity. And then the third and final upside we looked at was this. Are investors thinking, you know, diverse companies are simply more ethical investment targets, like the opposite of a sweatshop? So I'm not going to invest in a sweatshop no matter how profitable and diverse companies are? Kind of the opposite. This. This is the kind of place where I, I want to put my capital in. And so those are the three upsides we asked investors about. The three downsides we looked at were, do investors think that more diverse companies are going to suffer more from, like, negative stereotypes, like, maybe someone has a beef with women, not necessarily me, the investor, but like, maybe people won't buy from that company and as a result, I'm afraid of investing in that company. That's the one possible downside. And then the last two we looked at were, do investors fear that there's like, more interpersonal conflict among more diverse companies or more conflict over the tasks? Like, do people just fight about the best way to get things done? So the punchline is that investors have feelings about all those things, but only their feelings about the upsides track their investments. The three downsides track like 1% of variation in their investment behavior. The three upsides track about 99%. So overwhelmingly, you know, they have feelings about a lot of stuff, but they're willing to put their money where their mouth is for the upsides and not for the downsides. Okay, that's it. That's a good stopping point.
- [51:00]
I love it. And such good information for people to make the case for positive diversity. So thank you again, David, for joining me on the show. We will, of course have links to your website and to the paper and LinkedIn for people to connect with you. It's just been delightful to chat with you here today.
- [51:19]
Great. Yeah, thanks. And if you're interested in learning more, my website is www.DavidPDaniels.com.
- [51:28]
Wonderful. Thank you.
- [51:30]
Thank you.
- [51:31]
Melina Palmer
Thank you again to Dr. David Daniels for joining me on the show today. What got your brain buzzing in today's conversation? For me, I really love how this research shows the value in continuing to ask really great and thoughtful questions and knowing that even when something hasn't been able to be answered in the past, it doesn't mean that it can never be answered. And I particularly love that he was able to find an existing data set and look retroactively using modern modeling and tools to unlock new insights. Super fascinating stuff. So my advice for you is to keep asking thoughtful questions. Surprise, surprise.
- [52:08]
Dr. David Daniels
I know.
- [52:08]
Melina Palmer
Keep looking and digging and wondering and seeing what you can find. And know that while David's research is using very complex research methods and supercomputers, you don't need that to have interesting and important findings for your work, project, or company. The episode I refreshed this week was with my tips for setting up experiments in your business, including my three main rules to do your own brainy experiments, and for most of us that actually means keeping it small. So we have essentially opposite ends of the experiment spectrum covered on the podcast this week. If you haven't listened to that one.
- [52:44]
Dr. David Daniels
Already, I really recommend checking it out.
- [52:47]
Melina Palmer
And all these tests and experiments, they matter. I want to thank David for sticking with his big question until he was able to find this fascinating answer, which can hopefully lead to more thoughtful and diverse organizations who also see value on their bottom lines. Win, win wins all around. As we close out the show, don't forget about those show notes, which include links to my top related past episodes and books, ways to get in touch with David and myself, and more. It's all waiting for you in the app you're listening to and@the brainybusiness.com 477. And thank you again to Dr. David Daniels for joining me on the show today. It was a delight to chat with and learn from you. Join me Tuesday for another Brainy episode of the Brainy Business Podcast. It's going to be a lot of fun. You don't want to miss it. Until then, thanks again for listening and learning with me, and remember to be thoughtful.
- [53:46]
Thank you for listening to the Brainy Business Podcast. Melina offers virtual strategy sessions, workshops and other services to help businesses be more brain friendly. For more free resources, visit thebrainybusiness.com.