Maroš Servátka is a Professor of Economics at Macquarie Graduate School of Management, which is Australia’s leading graduate business school. Maroš served as the Economics Discipline Leader at MGSM in 2015-2018. He is the Founding Director of the MGSM Vernon L. Smith Experimental Economics Laboratory and the Founder and past President of the Slovak Economic Association. He has been awarded national and international competitive grants and his research has been published in leading economics and management journals.
Maroš: I grew up in Slovakia under communism, and one of the things that communists weren't able to do. I mean, obviously, there was the Iron Curtain. You couldn't get through the borders that kept us locked in. But my parents lived in Bratislava, the capital of Moscow. Like, yeah, well, back then it was still Czechoslovakia. And so we were the Slovakian Republic. And Bratislava is right on the border with Austria-Hungary now, one thing that communists couldn't stop was the TV signal getting across the border. And so as a teenager, my favourite thing to watch on a black and white TV was Austrian commercials, and I was watching them and our scratching my head wondering how come they have different, you know, more and different types of chocolate? Whereas we had one and it tasted awful. And that sort of, let me, you know, to thinking about economics, about markets. Communism was really known for the, you know, centralized economy.
You had the central planner and I was just really fascinated as a teenager by how come some countries do better in other countries do worse. And so that actually led me to study economics. And then fast forward into the nineties, my father was a diplomat. We lived in Poland and so I started doing my masters in economics at Warsaw School of Economics, which was a really good business school in that part of the world. And I met a small group of like-minded people who were just fascinated by economics and by economic theory. And so we started reading together. Now, mind you, that time at that time, there weren't that many economists in that part of the world, right?
Because a lot of them were just trained in the sort of the Marxism-Leninism that I think. But we had some really fantastic mathematicians who were supporting us. And so we just grabbed textbooks from, you know, Western textbooks about economics and sort of modern neoclassical economics. And we started from that and mathematicians were just helping us get through these models and understand the logic of it. And I just absolutely fell in love with economic theory because it was explaining everyday behaviour interest. It just made sense to me. And then I just decided that I wanted to be a scientist. I actually wanted to do the research on my own. I started the PhD program and part of the PhD program. We had a class that was called research methods and economics, and that we were supposed to write an essay. And the essay was about, Well, how do you conduct research?
You formulate the theory and then you test it because that's the way of how we progress in science and how we increase our understanding of the phenomena. And but I thought like, Look, I can actually take this opportunity to learn something new. And so I went we already had internet right in the nineties. Now there was no Google Scholar. There was no Google search. And you might remember Netscape Explorer back then. And I came across these essays by certain Vernon Smith, who started explaining that economics is an experimental science. I'm like it was considered
to be kind of like observational, kind of like astronomy, right? That that we formulate theories and then we wait years and years for the stars to move. Or something happened in the economy to test whether our predictions were correct or not. And Vernon said, Look, economics, it's not that economics is just like hard sciences.
We can conduct experiments. And I just thought this was absolutely fascinating. So I went to my advisor and I said, Look at the models that I'm building. I really wanted to know whether they work. I wanted to test them experimentally. And she said, I have no idea what you're talking about, but if this is something that you want to do it, perhaps you should go to the states and to continue your PhD there.
And so I applied to the University of Arizona, where Vernon was. And it was the meccaof experimental economics back then. And they accepted me to be in a PhD program, and I just started conducting experiments basically, since. Um, a very important thing that happened shortly afterwards was that Vernon won the Nobel Prize for Economics and for introducing experimental methods in. You into economics and experimentation became experimental methods became mainstream. And that was also important because other researchers started using them more and more. And we also communicate with the business environment, and so we have been also trying to as a, you know, as experimentalists, we have been trying to share this knowledge because it's also practical knowledge that is very important and businesses can learn a lot from conducting experiments. So I ended up after finishing PhD, I did a short postdoc in Germany in Mannheim. And then after that, we came to New Zealand, where I was for nine years at the University of Canterbury. And then after that, I got headhunted by Macquarie Graduate School of Management in Sydney. Now it's Macquarie Business School and one of the things that I do over there is I work with as I teach MBA students and we work with them and with their businesses and their organizations trying to figure out what works and what doesn't work. And obviously, our relationship with New Zealand didn’t end. I mean, we love New Zealand and we have been sort of coming to New Zealand all the time, and it's just a beautiful part of the world. We absolutely love New Zealand, so that sort of gives you some perspective of where I am and why I'm here.
Maroš: So one thing that really surprised me the most about the graduate program in the US was that up until then, I was pretty much a recipient of knowledge, and I sort of thought that the PhD program is going to be about learning new stuff. And yes, it was the first year, but even in that first year, they said, OK, now we want you guys to start thinking about how you can actually push the boundaries of our understanding. How is it that you can come up with your own discoveries about what works and what doesn't?
And I wasn't really thinking in these categories at all. I was trying to understand how things operated using the existing theories, using the existing knowledge that we had. And this sort of perspective was really mind-blowing because from that point on, once I understood what real research is and what science's pages, you can almost see my dream walking. My wife laughs at me that whatever I see, I just start thinking about why does it work this way? And I'm sort of never satisfied because, you know, when we even when I answer a question, usually the answers, the answers they multiply in a sort of linear way. But once you find an answer, there's like five more questions new questions that you get the right answer to. The questions multiply exponentially. And so that was the beauty of the PhD program. And, you know, something that they taught me at Arizona that I'm incredibly thankful for.
Maroš: One thing that I often have, you know, deep discussions with my MBA students is that we can use data to understand why people do certain things, why businesses behave in a particular way. And the interesting thing is that people often think about correlations, right? So we, you know, businesses nowadays with the sort of digital platforms that they operate through have a wealth of data and then people try to look for correlations about, you know, two things that potentially could tell them about what's happening in the business.
The trick here is to understand that not all correlations tell us much, right? Because sometimes they could be just spurious. It could be just things that are absolutely coincidental. And the experiments that we run with organizations, with businesses or businesses that we can run ourselves can actually teach us about causality. Right? So is something really, truly causal? Or am I just seeing things by an accident? And this is very important, right? Because if you bet on correlation and it's only a correlation, not causation, that can cost you a lot of money. On the other hand, if as a business you conduct an experiment and you learn that something works, that's a great thing because then you can actually scale it up or you can try to scale it up, implement and potentially become a market leader. On the other hand, which is also important is that we need to understand that when we learned that something doesn't work, that that knowledge is equally important as learning what you know, what works.
I actually have a beautiful story about that if you're interested. It's probably about one of the first experiments ever conducted. So this happened to probably one of the first trials that has been ever conducted was by James Lind on a ship, Salisbury. And so back then, the major problem that ships had was scurvy, right, people would just sort of go off on of these journeys. And after several weeks, they would develop scurvy and the ship's Salisbury that James Lind was on. He noticed that some crew members started already developed symptoms, so he decided to run a first experiment, the first randomized controlled trial, and he split the crew into separate groups. And each group was administered what was known to be or considered to be a medicine for scurvy back then. And so he started with saltwater. There was one group that was getting another group that ended up getting an orange, another one and ended up getting a lemon. And he noticed that the ones who were getting oranges and lemons were getting better. And that basically led to the discovery that vitamin C, later on, is what helps. And so people take this as the moral of the story. Now what is equally important and what often we forget about is that in that same experiment, James Lind discovered that we should not treat scurvy with saltwater.
And so in a similar fashion, right, if if you are a business, you really want to know what does not work. Well, maybe if he was even more ahead of his time, he could have mixed the saltwater with lemon and created the ultimate fasting cleansing diet. Right? And again, experimentation, right? Yes, trying new things and sort of understanding how this works. Absolutely crucial. I don't know whether, you know, do you know what to do? Penicillin and the microwave have in common what tell me. So both of them are actually failed experiments. Both of them are failed attempts, right? So Alexander Fleming, when he was conducting his experiments, he got a little bit more than his petri dish. And then he discovered that the bacteria thing was Staphylococcus that he had there. Where actually not. They stopped growing. Right in that accidental discovery that led to the creation of penicillin, Percy Spencer was experimenting with these tubes with the vacuum. And then he was sort of pointing them out. And then he noticed that he had the chocolate on him and the chocolate melted. And so we started pointing the tubes again, and they figure out that actually, that was what caused it. And so then that discovery then later led to what we know today as a kitchen appliance called the microwave, right. So these failed attempts can also be a source of new knowledge. Now, that doesn't mean that every experiment obviously is going to end up with such a, you know, such a great discovery, but it's about trying new things and learning what works and what doesn't. If you can, I'd love to hear about some of the experiments that you have conducted and especially the economic experimentations that have been in place and what are some of the, well, hypotheses and findings.
So here is, I'll talk about something very recent that we have done, this is with my colleagues down at the University of Otago. Stephen Knowles and Trudy Sullivan and Rod Catch, we conducted an experiment testing procrastination and the effect of deadlines. Now, procrastinate is also mean procrastination. It's a problem, but probably all of us suffer from it.
We procrastinate on things all the time, and if you know, it really just depends what task it is that we procrastinate, right? And the reason why people procrastinate is that we tend to consider a few present to be more important in the future. And so when there's an onerous task that we need to complete. That onerous task has costs that are immediate. Right. But that the benefits of the task just come in the future. And so the benefits that we are getting get that discounted more heavily. And so it seems so real when is when I'm a procrastinator? I'm really faced with the following choice I can complete the task now and incur these costs right away. And they seem massive. And then I'm going to reap the benefits. But the benefits come in the future. Or I could postpone the completing the task, in which case the costs are also going to be postponed and they don't seem as big anymore. Hmm. And so many people actually do that. Now there are two types of procrastinators. There are procrastinators who are naive and who don't realize that tomorrow eventually will become today, and they're going to be faced with the very same problem. And then they postpone again and again, and they never complete the task. On the other hand, there are sophisticated procrastinators who realize that they have self-control problems. I mean, the evidence is vast, right? Sometimes it's just a little task that we have to do. Sometimes it's an important task like, you know, getting your super organized right? Refinance your mortgage. Be tax returns with tax returns.
If you don't complete them right there, penalties are severe and yet we still postpone these things. Even things in personal lives, right? We realize. So let's not talk about financial things. We realized that exercise is really important. But how do we feel like exercising today? I'm going to do it later, right? I should not be smoking, but it's really hard to quit. OK. And so I should go to see my dentist and get my teeth checked, right? I don't really feel like doing that today. And so, you know, we have a lot of evidence of this procrastination. So so the question is, well, how do we how can we actually deal with it? And what helps and I don't know, how about you? But deadlines organize my life. And so one of the things that we tested with Stephen, Moorad and Trudy was what if we change the deadlines are deadlines going to help procrastination? Now there has been some research on that had previously. But let me just sketch out the two underlying ideas behind why and how deadlines should work. So sort of the neoclassical the rational model tells us, the longer the deadline, the higher the probability that people will complete it. Right. Because you just have eventually you're going to find a moment in time when you are not as busy when your costs are a little bit lower, and so you will just do that. On the other hand, for procrastinators. If I give you more time, it just becomes easier for you to postpone this. Yes. Now, so for procrastinators, giving them more time can actually be detrimental. The question is which one of these effects prevails? Most previous studies actually just looked at two deadlines, and they said, Well, the completions, the longer the deadline, the lower the completions that we had. So that means by extending the deadline, that's a terrible idea, which is should have short deadlines. No one ever looked at it, Well, what if we don't give people deadlines? Hmm. How is that going to affect things? And again, so our thinking about the problem is when we, which is has been completely ignored by the previous research, is that when I give you a deadline and I ask you to do something, this could be a favour for me. Or this could be, you know, I want to say, Glenn, could you please complete the DIY project, right? Could you donate to, you know, the charity at sensitive request? I could. You could donate to our volunteering organization, ask you to come and help when asking and when giving deadlines that information that is contained in a deadline is also going to tell you about the importance of the task and its urgency. And so asking you to do something very quickly. Implies urgency, I say, look, I need you to get this done right away, could you please help me with that? And so that short, the deadline imposes urgency.
On the other hand, if I tell you, Glenn, could you please do this for me? But look, there is no rush. I'm not going to even touch it for the next month. I've just basically given you permission to procrastinate. It's a free pass. Exactly. And so I'm signalling to you that this is not important. And because we are humans, we also our memory is imperfect. We also tend to forget. And so that also what often happens right, tasks with long deadlines to not get completed. But now here's an important question. What if there is no deadline? And you know, the standard behavioral answer would be, well, if there's no deadline, this is like, you know, the infinite deadline, really. And so the completions should be the lowest. Now notice, however, that when charitable organizations, volunteer organizations ask for help. They don't often specify deadlines. And when they do that, they do that because not seeing that there is a deadline, basically. Implies the urgency you want you to help as soon as you can right away. Right. This is and this is really important to us. And so we tested that. So we ran an experiment with a representative sample of the New Zealand population. We send out about over 3000 letters to people all over. New Zealand was stratified across genders, across different age groups, and we asked them to fill out a short ten-minute online survey. And for completing the survey, we would pay $10 to a charitable organization of their choice. And we had three treatments for three conditions, one with a very short one-week deadline another one with a one-month deadline. And the third one did not have a deadline at all.
We didn't even mention it in the letter, and the results were very surprising. The short deadline elicited a really high response rate when we went to them for a one-month deadline. The response rate dropped, but then when we came to the no deadline condition, the response rate was the highest. And so we find this sort of non-monotonic effect, right? Which again, was not predicted by any previous theory, no one really. And the reason why we see this, we could also look into the completion and how how quickly people responded in the no deadline and a short deadline, people responded immediately. Right. They did it within the first three days, the vast majority of those who responded, whereas, in the one-month deadline, many people just postponed. Right. And this is precise gets to the issue of the deadline. Can signal urgency tells you that the task is important. Which people in which case people respond right away. If you, however, give them permission to procrastinate. They are going to do so, and then some of them might eventually also forget.
Maroš: So my first observation is that we just don't experiment enough in general businesses are quite reluctant to try new approaches. Why? Well, first of all, many managers hold the belief that the current approach is best. And it's it is, you know, working somewhat. I guess that's why you have the business still operating. But they fail to recognize that often the way we currently do things as an organization or as a business is due to historical circumstances. It could be due to precedent. And also, we as people often have this status quo bias. We don't like to change things, right? Sometimes managers can. Also, they're known to be overconfident and that things are just going to work quite well.
And so these forces cannot hold the old approach right together. And there's this, like I said, to the status quo and people are just reluctant to change things. The second problem in an organization is the fear that new approach might fail. And you can understand how, for example, a manager might not really want to talk to the CEO, to the board and explaining to them that whatever he or she came up with didn't work. We see a similar thing with politicians, right? They also prefer a status quo, because what if? What if things don't work out the way we plan, right? Then you have to explain himself. So several, several years ago now, when we still lived in Christchurch when I was at University of Canterbury, I applied for a large grant and we actually didn't get it. And it was about how to what we were proposing to test how to decrease insurance fraud. Hmm. Then about a year, year and a half later, after our unsuccessful grant, I met at some reception and gentlemen and we started chatting and eventually said, You know, I recognize your name from somewhere and as it turned out, he was on the panel. He was one of the judges of the proposals that we put in. I said we thought it was really fascinating because you were sort of asking, Well, how would you decrease the insurance problem, which is one of the biggest problems in the insurance industry? And our question just was, well, how are you going to get the data?
So well, we created the data, right? We run these experiments to see what works, what doesn't. And he said, well, there was one problem. And that is there are all these solutions that you economists came up with. But how do we know that they worked? How do we know that something that has not been done before? Is actually going to pan out. That's exactly the issue. Well, if we never try things, we're never going to find out. So yes, new things might fail, but we actually have to test whether it works. And what is important to realize is that trying is not the same thing as implementing. And so you see successful businesses around the world experimenting with new approaches, testing new things. Right. So Facebook has multiple versions when they test two different interfaces. Google would test the auction format of the auction, so that happened in the background when you put in a search word. Uber has a platform that is designed to test what works and what doesn't. In fact, they encourage their employees to log into the platform and suggest experiments to improve the performance and also suggest how to evaluate what works and what doesn't. So at any given time, they have close to 1000 experiments running, right
Maroš: Absolutely. The important thing about experimentation is that it's a very important source of knowledge. And well, on one hand, we tend to be very happy when something works and we figure and figure out, too, that this new approach is successful. That's a great thing. But notice that when it comes to learning itself, we only kind of got a confirmation about what we already knew because the reason why we tested this approach was because we were thinking that it would work. And so the actual knowledge that we generate is from finding when things don't work. This is, I'll put it in sort of my scientific perspective. The most I learn is when I find out that I was wrong because that challenges my thinking further. And I have to figure out why is it that I was wrong? What was what? What exactly went wrong? And then allows me then to improve my service, improve my product, make it more desirable to my customers and whatnot. So it's it's these little failures that generate even more knowledge than the little successes, not always be. The little successes are great because then we can use them for our business. But in terms of learning itself, right? Like I said before, finding out what doesn't work is equally, if not even more important, than finding what works.
Maroš: Precisely why? And, you know, the practical aspect of things is, OK, experimentation is costly and time-consuming. So why would we actually want to engage in this? Well, first of all, let me change the question and let's ask what is the cost of not experimenting? What is the cost of not trying new things? And if you don't try new things, if you don't innovate, if you are not proactive, that can lead to lost opportunities to become a market leader, your market share can decrease. There's a really nice disaster story by Netflix. I forget what year this was. They changed their pricing system. It was about a decade ago, and they implemented it. For all of their customers and customers just hated it, and B, they were able to recover, however, do they, for example, tested this in the San Diego area? They would have found out that their customers didn't like it and they would have actually stopped from scaling it up. And so what I'm trying to illustrate here is that if a new policy is implemented and it's not working, the cost can be astronomical. Yes, but the cost of experimenting are fixed and limited and could be budgeted for. And so that way you actually help your business grow.
You understand that the environment in which you are operating better, you understand your customers better. And those are the advantages of experimentation of things that you can that you can actually learn. And what is important at this experimentation allows you to stay ahead of the curve. It is not one of experiments that you run. In fact, experimentation allows you to learn continuously and recently collaborated with the National Transfusion Service back in Slovakia. And we were designing a blood registry for them, basically trying to figure out how to make the blood collection system more efficient. And when we started this discussion, they again, it's an organization that is not naturally prone to experimentation, and they said, Well, OK, so we're going to implement this blood registry and then what do you finally leave us alone? And I said, no, because what we are trying to do is to learn continuously.
So once we find out that something works, we're going to implement that. If we point out that something doesn't work, we're going to try something different, a new thing. And so this is it's an example of how these little steps in business in, you know, charitable organization and nonprofit organization can lead to better service.
Maroš: Exactly. And this is where the experimental economics approach is very important because we look at what people do, not what they say. So, economists, we're quite skeptical about surveys because, in surveys, there are other influences that might matter. So suppose we had a conversation and I wanted to study generosity and I would ask you, Glenn, could you tell me, are you a generous person? And I don't know, how about you? But if somebody asked me that question, I would really be concerned about my image, and I would try to say, of course, I'm a very generous person. But then you're not really learning anything. So you wanted to quantify things. And so I'm going to give you a hypothetical question now. And to demonstrate that hypothetical question, hypothetical scenarios also do not always lead to truthful answers. So if I asked you, I suppose, Glenn, you won a million dollars on a lottery, how much money would you donate to a charity?
Maroš: Yeah. And so that was actually quite an honest answer. And I'll put it again in my own perspective, If I got asked this question, I would say I would donate to the entire million. Why? Because I don't have it this. It doesn't cost me anything. And does did the charity get anything out of it? They didn't. And so we need really need to look at what is it that people do as opposed to what they say they will do. And this is also common with procrastination what we get asked about. Not many people will tell you, Oh, I'm going to start exercising. I mean, we're almost we are. It's almost Christmas. But people are going to start with their resolutions again, and many will buy a gym membership because they're going to say they're exercising. And those are their intentions. Now, fast forward to say February, how many of them are in the gym? How many of them changed their diets? How many of them stopped smoking? And so what people say that they will do is not necessarily what they truly do. And as economists, we're interested in the outcomes. So in all of these experiments that we have discussed, we measure, we look at behaviour, we look at functions, we measure how many people have donated blood. How many people volunteered. How many people donated to a charity, and how much.
Maroš: I haven't conducted any research on my own on this, but this is from what I've read in the literature when it comes to unsolicited feedback. There are two types of people who are going to provide feedback, and those are the polar opposites, the ones who are incredibly happy with the service and those who are extremely unhappy, like the who just hate to do the service. They didn't like the product at the end, and they're just really grumpy about it. The people who received the product and the product was OK are probably the ones who are just not going to bother providing feedback. When it comes to solicited feedback, this is the type of selection problem that we often deal with surveys, so,
If let's think of a scenario, you get a phone call from a company conducting a survey. Right, you're busy. And I said, Glenn, we would like to ask you 250 questions. It probably takes 45 minutes or more. Would you be available? And the answer is likely going to be no. And so the only people who are going to be willing to answer these questions are those who are not busy or who have nothing better to do at that moment, which is then going to lead to a selection. Meaning if you are not getting a representative sample, you're not getting, you know, sort of a nice cross-section of the population, but you are just getting people who you know from particular in a particular age group or the particular socioeconomic status that are going to be answering. And so then at the inferences that you can draw from these surveys are just not going to be extremely valid. The experiments that I talk about the particularly the field experiments that we have conducted, the huge advantage is that people don't even know that they're an experiment. And therefore, there is no issue with scrutiny like the ones that we sort of discussed. I want to look like the nice guy. They are not really biased because they don't anticipate being an experiment and they just go about their everyday business just like they would otherwise. And the researchers are able to collect data on their behaviour, which then leads to allows us to test the underlying theories and then uncover the relationships of interest.
Maroš: I'm going to finish it on not experimenting enough, so we don't experiment enough with our customers, we don't experiment enough with our employees, we don't experiment enough in our own personal lives. And I'm saying this as a professor who teaches about experimentation, and I often catch myself that I just do things in and we are just creatures of habit. I don't do things in a particular way. And how basically experimentation is both a skill, but it's also a conscious decision. And the more we get into experimentation, the more we get into trying new things, creating perturbations in our business lives, in our personal lives, the more we're going to learn and through observing what happens and then we reflect on what happened and why we behaved in a particular way, or why our business performed in a particular way, why our customers did this as opposed to something else. And that reflection is then going to allow us to formulate new ideas and we can't then again, test.
Where to find Maroš Servátka:
Website: https://sites.google.com/site/marosservatka/
LinkedIn: https://www.linkedin.com/in/maroš-servátka-7122294/
If you want to understand your customer feedback better, talk to the team at Touchpoint Group.
Fill out your details in our contact form and we’ll help answer your questions!