00:00:00 Chapter four asks supply chain’s true goal
00:04:45 Non-aggression frames voluntary, money-making cooperation
00:09:30 Regulation limits differ from forced transactions
00:14:10 Sustainability follows customer preferences, not separate goals
00:18:00 Economic laws explain supply chain boundaries
00:22:10 Dollars become atoms, then more dollars
00:27:10 Mainstream misses time-sensitive rate of return
00:32:10 Service levels fail as economic objectives
00:37:00 Aviation retrofits reveal proxy limitations
00:41:00 Aperiodic returns add operational time granularity
00:46:45 Spoon opportunity illustrates peak-return cutoff
00:51:45 Service levels ignore timing and opportunity buys
00:56:45 Proxies force waste and miss windfalls
01:00:10 Shadow valuations capture off-ledger future effects
01:05:10 Rare stockouts require thought experiments
01:10:20 Diapers, insulin, and customer lifetime losses
01:15:20 Four chapters now enable profitable reinvention
01:19:00 Double-entry accounting unlocked capitalism’s scorekeeping
Summary
Supply chain, in this discussion, is not about hitting fashionable KPIs or moral posturing. It is about making money by allocating scarce resources better over time. Joannes argues that supply chain belongs to applied economics: firms exist within webs of specialization, and decisions should be judged by rate of return, not crude proxies like service levels. Time matters because a profit earned quickly is not the same as one earned slowly. He also defends “shadow valuations” as a way to account for future consequences—lost trust, customer attrition, strategic damage—that accounting ledgers do not capture directly.
Extended Summary
The discussion centers on a deceptively simple proposition: the purpose of supply chain is to make money. Not “to serve many stakeholders” in the abstract, not to optimize fashionable metrics, and not to pursue a vague moral mission, but to convert resources into more value than they started with. Joannes argues that many confusions in supply chain begin when people evade this point and replace it with softer language that sounds noble but provides no clear standard for action.
From there, the conversation places supply chain within economics rather than mathematics or managerial ritual. Supply chain exists because of specialization and exchange. The Ricardian logic of comparative advantage explains why firms do not do everything themselves, even when one company appears superior to another in many respects. The boundaries of the firm are not arbitrary. They emerge from economic reality.
The central practical concept is rate of return. It is not enough to turn $100 into $110. The relevant question is: how long did it take? Time is not a side issue; it is the issue. Supply chain is an ongoing, repeated process of turning money into inventory and operations, and then back into more money. A slower gain may be inferior to a faster one, even if the nominal gain looks similar. This is why Joannes rejects the common habit of judging decisions by static targets such as service levels or safety stock formulas. Such measures are, at best, rough proxies. At worst, they push firms to tie up resources unprofitably or miss lucrative opportunities entirely.
This leads to his criticism of mainstream supply chain thinking. In his view, it is dominated by cost minimization under arbitrary constraints, rather than by maximizing economic return. Service level, in particular, is criticized as a simplistic percentage that often misses what customers actually value. Customers do not buy percentages. They buy outcomes, quality, timeliness, relevance, and trust.
Finally, the conversation introduces “shadow valuations”: economic judgments assigned to important future consequences that do not yet appear in the ledger. Customer attrition, damaged trust, bad habits created by discounts, or catastrophic stockouts in critical items may not be directly measurable from historical records. Yet they are economically real. If managers ignore them because they are hard to quantify, they do not become less important; they merely become unmanaged.
The broader message is that supply chain should be judged by whether it allocates scarce resources in ways that compound value over time. Everything else is secondary.
Full Transcript
Conor Doherty: Welcome back. This is episode four of a special series where Joannes and I go chapter by chapter through his new book, Introduction to Supply Chain. We go back and forth. We debate the merits of the ideas, particularly from the perspective of someone who might be a first-time reader who’s never heard of Lokad, has received what you might call the mainstream approach to training, and they might have certain questions about the content. That’s the position I take in this series.
If you haven’t seen the first three episodes, I strongly recommend that you watch them because, as you can imagine, this is an ongoing series, an ongoing discussion, and from time to time we do make callbacks to things that were discussed previously. And with that, Joannes, thank you for joining me. Episode four, we will be discussing chapter four, titled The Empire Strikes Economics. Economics. So, Joannes, what is the grand central thesis of the chapter called “Economics”?
Joannes Vermorel: It’s about answering a very, very simple question: what are we even trying to do? And then the consequence of that.
Conor Doherty: So, what are we even trying to do? Making money. Okay. That’s the word. You’re very explicit about that. Unapologetically, capital business.
Joannes Vermorel: Yep, exactly. We are going to make money. And, by the way, it has consequences. We are just making money. That’s it. So it is not that we need to make money. It is, in fact, just what we should actually do, and that’s also the part because people can get confused. They say, “Oh yes, we need to make money and plenty of other things.” And I say, no, we need to make money, period.
This is important because when people think, “Oh, but if you’re just making money, period, you’re such a bad person,” et cetera, it reflects that people don’t understand what making money actually means. If they don’t understand what making money means, then all sorts of bizarre consequences and bizarre misunderstandings unfold from that, and we drift into, I would say, irrelevant questions.
Fundamentally, this is very important, and this chapter, “Economics,” follows an epistemology where we have tried to eliminate, I would say, categorical thinking errors when we approach supply chains. Here again, we have eliminated the idea that we should think of supply chain as math, as solving math puzzles. This is not it. And we should not think of supply chain as a management thing where we are just slicing and dicing the division of labor. Again, slicing and dicing the division of labor is a very secondary concern. The primary concern is making money.
That’s the point of economics: to understand how this concern, “we want to make money,” relates to supply chain. And, in the grand scheme of things, if we are looking at supply chain in the grand tree of human knowledge, where do we place supply chain in this grand tree? That’s essentially the answer that this chapter provides.
Conor Doherty: Well, I have prepared questions, but you’ve already said things I want to follow up on. Again, because I know in the chapter you do make the point that you don’t want people to confuse supply chain with some sort of moral pursuit.
You talk about this in your own terminology. You talk about how there’s no moral dimension to what we’re doing here. We’re making money. But you just made the claim that most people don’t actually know what making money actually means. So, from the perspective of the book—which is, again, Introduction to Supply Chain—what do you mean by that? How do the ten million average practitioners not really understand what making money actually means?
Joannes Vermorel: First, I am putting myself philosophically in… what is my philosophical statement? I’m saying this is a game played for money that I will play without coercion. That’s it.
Nobody can actually physically force anybody to do anything. That’s the only thing that I’m saying. In real life, this is not true. In real life, if the French police want to take me away and throw me in jail, for good reasons or bad reasons, they definitively can. So, the use of force in real life is a real thing. You can get what you want by force. Governments do that all the time.
Conor Doherty: Yeah, but there are regulations that you have to—
Joannes Vermorel: I’m not saying otherwise. Force exists, and this is a dimension that is essential to human existence. But what I’m saying is that, for supply chain, for the context, we are operating in a paradigm of voluntary cooperation. I am making that clear. That’s why I say supply chains for armies and armed forces do not abide by the same code.
Conor Doherty: Yes.
Joannes Vermorel: They do not, because the use of force is on the table. Here, I’m just saying, okay, if we admit, if we start from the point that we are playing a game that is fundamentally cooperative, I cannot force anybody to do anything. My employees, if they don’t want to stay, they will leave. My clients, if they don’t want to stay, they will leave. If my suppliers don’t like me because I’m such a pain, they will leave. So, you see, nobody can be forced.
And when you say, “Oh, money can force people,” no, no, no. Money cannot force. You have, by the way, an expression, you know, “having fuck-you money.”
Conor Doherty: We may have to blur that on YouTube, but yes.
Joannes Vermorel: But you see, philosophically, I’m just saying that the game of supply chain is played from a non-aggressive perspective. That’s it.
Conor Doherty: Okay.
Joannes Vermorel: So, that’s the rule of the game. The only thing that I’m saying is that, for people playing supply chain, you’re not allowed to use force and guns. That’s what I’m saying. I think, for everybody, it’s self-evident, but I just want to point out that it is a very, very limited assumption. I’m not making a grand assumption. I’m just saying you cannot force someone with a gun to buy your product and pay you. That’s it. It’s a very basic expectation, and I think nobody is even thinking the opposite.
Now let’s bring it back to the initial case.
Conor Doherty: Yes, exactly.
Joannes Vermorel: When people say making money is a bad thing, that statement assumes that bad consequences will unfold, but that contradicts what I just said, which is that you can’t force anybody to do anything. Thus, for example, when people say, “Oh, if we let a CEO do their own thing, this CEO will do such a disservice to society, he will abuse customers, he will do this bad thing and that bad thing,” et cetera—how is it exactly going to work out?
If the CEO of a very large corporation starts to act in ways that are morally reprehensible, even if it’s legal, his own customers are going to leave. They are definitely going to leave. If the top management of Apple lose their mind and the next iPhone is a piece of crap, people will leave. Apple cannot force anybody to buy it.
Conor Doherty: Maybe people are in love with Apple and they will be so incredibly disappointed if the next iPhone is so bad.
Joannes Vermorel: But ultimately, Apple cannot force anybody to buy it. That is what I’m saying. Now, the problem is that we don’t live in a world where we have non-aggression. There are plenty of large companies that can definitely force clients to buy. Why? Because, in fact, they leverage the force of government to get their business.
In this case, we have plenty of—France is a specialist—so many… very few countries are purely free markets.
Conor Doherty: You touch on free markets and both—but again—
Joannes Vermorel: Again, the problem is that if we go into non-free markets, we are digressing into what is the appropriate use of force to get your way and make money.
Conor Doherty: But a lot of the companies, again, just to really bring it back to the concrete, a lot of the companies that we even deal with operate in heavily regulated markets.
Joannes Vermorel: Again, there are degrees. There is a degree where regulation just says, “Here are things that you cannot do, but you’re free to do whatever you want. I will not help you. I’m just saying what you cannot do.” I would say that’s our clients. So, they are regulated only in the sense that there are things that are not allowed. That’s it.
For example, if you’re a fashion retailer, the type of inks that you use for your clothes are extremely limited. You cannot use any kind of ink because some inks are very toxic. So it is heavily regulated. But that’s it. Nobody is forced.
Now, you have other companies—we don’t have those companies as clients—but other companies where the state says, “You have a mandate. You’re a private company, but I give you a mandate, and every French citizen is a mandatory client of yours. They don’t have any choice. They must pay.” So, you are essentially a private entity, but you have a mandate, and then you collect money from citizens. And if a citizen says, “I don’t really like you, you’re a private company, I want to quit you,” then the company says, “You know what? I have a mandate. It’s not an option. You don’t have this option.”
So here, if you don’t pay, it’s police and jail for you. You see, here it is a game played with force. We have to be aware that, in regulation, there are two completely different things at play. There is regulation that just limits what the company can do, but ultimately, as a customer, it’s still completely non-aggressive. You can still walk away. It just means that the company cannot do anything that you can dream up.
For example, if you say, “Oh, I would like a company to sell me a nuclear device,” and they say, “Hold up. No, no. Nobody, no company, will sell you a nuclear device. Sorry.” Fundamentally, that limits you as a customer and that limits the company, but still, the relationship of non-aggression stays intact. It’s just that there will not be transactions for components of nuclear weapons happening freely on the market.
So that’s the limit that we have, just like if an ink is extremely toxic, the state will not let someone sell that, even if the customer is willing to buy. Again, that’s just saying, “No, you’re not going to sell that, and no, you’re not going to buy that.” That’s a very small limit versus, “You company will sell something to this guy, and this guy can’t say no, and you can’t say no either.” So you see, it is force transactions.
What I’m saying is that the paradigm in which I operate for this Introduction to Supply Chain is essentially a non-aggressive paradigm. I’m just describing idealized supply chains where aggression is not on the table. That’s it. I’m not saying that aggression doesn’t exist in the world. I’m not saying that armies do not exist. I’m just saying that the playbook that is there—
Conor Doherty: Yes.
Joannes Vermorel: —is essentially non-aggressive. That’s it. That’s the only thing.
Conor Doherty: Okay. Well, again, there are a few points where I… because I don’t want to get lost. Look, I studied philosophy, you know. I don’t want to get lost.
Joannes Vermorel: Yeah, but it’s important because, if we want to understand why making money is not a bad thing, people—whenever they think this company is making money in bad ways—it is invariably because they don’t understand that the only way for… either the company is like a good player, so the company is literally doing what people want.
For example, maybe the company doing Pokémon. You think that Pokémon are stupid. That’s your opinion. And so you think, “Oh, the Japanese company selling those Pokémon, they should be banned.” But it turns out that there are millions of fans of Pokémon, and so they love Pokémon. All those people disagree with you. So even if you think that the CEO of the company producing Pokémon is doing a disservice because, in your opinion, it’s making people stupid, that’s your opinion.
In your opinion, the money earned by the CEO is unduly acquired. This is a categorical mistake. The CEO is just doing what people want. People want Pokémon. He gives them Pokémon. It is not for him to decide that, after all, Pokémon might not be a good thing for you, so he’s going to stop doing that. You see? It’s very different from the CEO making money because, for example, he is using the state to ransom citizens.
What I’m saying is that people are systematically conflating those things. If you approach supply chain and money and economics right—which is non-aggression—then I’m saying there is no such thing as people, out of voluntary cooperation, making bad money, because the money is voluntarily given in my paradigm.
Conor Doherty: I don’t think many people would disagree with that because, again, the thing is—the book is… you’re not writing to the general audience. You’re writing to the general supply chain practitioner audience who operate within the framework you’re talking about.
Joannes Vermorel: Yes, but you see, yes and no. The reason why I was making this point is that many of the supply chain books that I’ve read are failing on this test. They misunderstand this very point that we just described, and thus they end up with an incredibly fuzzy… when it comes to the goals, they end up with an incredible list of so many goals for supply chain. They end up with twenty different goals.
They would say, for example, sustainability. For me, that would be an example of why I do not think this is a goal. Sustainability is not a goal. The answer is very simple: people generally don’t want their environment to be destroyed. There are very few people who say—it probably exists, but it’s pathological—that they want to see the world burn, their forests burn, their environment destroyed forever, et cetera. It’s like 0.01% of the population. So 99% of the population want something that will be preserved. They want their children to inherit an earth that is better than the one they have.
So, how can you profitably operate anything if you have something that is so important for your customers? It’s not possible. And if we go to a real-life example, IKEA started to plant trees to build their furniture almost a century before these sorts of concerns came onto the table. We didn’t have to invent sustainability in the modern term for a company that is in the business of cutting forests to build furniture to realize that they need to plant forests. They started to plant forests.
People can confirm online. Off the top of my memory, I think they started to plant forests in the forties or fifties. They did that way before they became the giant they are today, where planting forests is an actual concern because they are cutting so much. But what I’m saying is that if you make money right, because people want something that will last… again, very few people have this sort of scorched-earth thinking.
So, “making money needs to be sustainable” is built in. What I’m saying is that if you expand the list of goals into cherry-picking plenty of stuff for virtue signaling, you are losing sight of what matters, which is making money, and that’s not a good thing.
Conor Doherty: Okay. Well, for the sake of pushing forward, I’m going to presume that most people who would pick up this book and read it would probably concede that point with you from the outset, if they’re operating within the same framework as you are, which is: if you’re working within supply chain to one degree or another, you’re assenting to operate within a somewhat capitalistic framework.
But I don’t want to get lost on that, because I actually want to focus on the actual meat of the economics argument. Because the whole thesis of what I know to be Lokad’s perspective and your position as you present it in the book is that supply chain is a branch of applied economics. In the book—this is chapter four, so we’re now well into the book—you start listing general laws, as you call them, general laws of economics that cannot be evaded. You give three core examples: obviously supply and demand, most people will be familiar with that; diminishing returns, most people I think might be familiar with that, some people might not be; and then the Ricardian law of comparative advantage.
Now, again, that’s in the opening sections of this chapter. From the perspective of the supply chain practitioner, why are those three listed as core concepts that you must grasp? Supply and demand goes without saying, but things like diminishing returns and the Ricardian law of comparative advantage—how does that fit into the average demand planner’s day-to-day?
Joannes Vermorel: On a very, very grand scale, supply chain—supply chains, not the chains generally I’m describing, like the cohorts of companies—is different. Supply chain is the science of the allocation of resources. Supply chains, plural, are much more like a macroeconomic perspective, with a web of interdependent companies.
So the question is: why do we even have this web of interdependent companies called supply chains, plural?
Conor Doherty: Yes.
Joannes Vermorel: Why is that? The short answer is division of labor. It makes everything better when we do that, and we make more money.
Conor Doherty: So, and we make more money. Exactly.
Joannes Vermorel: That’s very good. So it’s something that is win-win. Everybody wins. But the question is strange. You have to pause for a second. How come, in this supply chain, we have companies that seem so incredibly backward? If there is a company, let’s say Amazon, that is better in every dimension—every single dimension—Amazon is better than company X—
Conor Doherty: Yeah.
Joannes Vermorel: Why would Amazon ever rely on company X? It’s a very strange question. Why should Amazon not internalize what company X is doing if Amazon is better on every dimension?
The short answer to this question is the Ricardian law of comparative advantage. It explains why, even if you have a company that is worse on every dimension than another company, it can still be profitable for both to trade. Again, that is a fundamental understanding, because that’s how you will understand the boundary of your company. Your company is a piece of the supply chain; it’s not all of the supply chain.
You do not go from extracting minerals from the earth to essentially selling car leases. You see, there is a whole chain of things between “I extract the metal” and “I am a financial operator selling not even the car but the lease of the car.” There is a whole chain of things. I’m just saying, to understand why those boundaries even exist and emerge, and thus to understand the limits of the game that you’re playing—because you, as a supply chain practitioner, are operating within a single company—you need to understand this Ricardian principle. Otherwise, you will not understand why your company is not doing everything end to end.
Conor Doherty: Well, this actually sets up the concept which I think is core to understanding the chapter, your overall philosophy, and I would say what Lokad does itself, which is: what is the return on the allocation of your resources?
So, again, if you’re talking about the example of Amazon and company X, Amazon can do all of these things, but the whole point is: where does it maximize the return on its time, energy, and resources? Even if the other company produces an inferior version, if it’s still more profitable for Amazon to outsource it to that company, that should be the choice, okay, because you’re trying to maximize what you call your rate of return.
So that, to me, is the central concept. We can get into decisions in chapter eight, but how we even rank decisions is predicated upon an understanding of the concept of rate of return. So, Joannes, explain to people who have not read the book what that is. Actually, explain it to me as if I’m a busy ops director and I just need to know: okay, I need to know more about the financial side of my decisions. So, we say making money. Let’s unpack that.
Joannes Vermorel: Exactly. Making money in supply chain means you take your dollars, you convert them to atoms. That’s going to be stuff, physical stuff.
Conor Doherty: I’m a busy ops director.
Joannes Vermorel: And you do things like transport, transform, distribute, advertise, and at the end of the game, magic happens. The stuff, transformed, transported, et cetera, is converted back into dollars.
Conor Doherty: Yes. Hopefully more dollars.
Joannes Vermorel: More dollars. Thus, you had a starting point of dollars. You had, like, a hundred dollars initially. You convert that to stuff. You do stuff to the stuff. At the end, you transform the physical stuff into dollars back, and now you have a hundred and ten dollars.
Conor Doherty: Nice. Money.
Joannes Vermorel: Money. Okay. Now we have actually increased the amount of dollars. We are making money. Now, the idea of rate of return is that we want to play that game as fast as possible. How long did it take to go from dollars to stuff and then stuff to dollars again?
Conor Doherty: Yes.
Joannes Vermorel: If I can do that in one year—one hundred dollars in January invested, and by the end of December I have one hundred and ten—I have a 10% annual return. If I can do that in one week, I have 10% per week, which, if I annualize that, if I compound, is going to be enormous.
So it’s not just investing and getting more. It’s about doing it. It’s time-sensitive. It is doing that in the timeframe that is as short as possible.
Conor Doherty: This is where we’re getting, I think, into the philosophy, but when we’re getting into the concrete topics, like time—
Joannes Vermorel: Yes.
Conor Doherty: Because, again, no one will challenge you about the idea of, “Oh well, yes, if I invest one hundred, it would be nice if I made more than that.” I mean, that is the absolute basis not only of supply chain, but everyone’s own time. I do things, I want to earn money for the time that I have invested.
But when we talk about time, that to me is a concept that is a little fuzzy to people, and that gets into a periodic rate of return. So it’s not just rate of return. Please expand a little bit more on why time preference is important here.
Joannes Vermorel: That’s the key. In the supply chain context, making money is something that is fundamentally time-dependent. You cannot remove the time dimension; otherwise, you don’t know what you’re talking about. So when people say, “I earned a million. Oh great, I earned a million.” The question is: yes, but how much time did it take you? Because if it took you a lifetime to get there, you might not be that rich. If you divide that by forty years, it’s fifty thousand dollars per year. It’s nice, but you’re not rich. If you say, “I made a million, but that’s what I did last week, and I’m going to do that again every week,” then you say, “Oh, we are not in the same business.”
Supply chain is an iterated game. It’s not winning the lottery once. It’s doing it over and over and over, and as fast as possible. Thus, the time dimension is fundamental. It’s literally: how much money can I make as fast as possible? And the rate of return gives you something that says, if I have this quantity of initial money, after a year, if I annualize, how much will I have? What we want to say is that we want to allocate our resources, every allocation, so that we maximize this rate of return. So that, when we look ahead into the future, we have the thing that is compounding as fast as possible.
This is why, again, capitalism is so incredible. It is trying to have this initial seed of value compound into something bigger. That’s how all the very, very large fortunes happened. By the way, that’s why I said force and not force is completely important. Pre-capitalism, the only way to amass a fortune was to have an army and conquer. That was the only way. Expansion, grabbing, pillage—that was the game.
If you look at all the people who were very rich in antiquity, with a few exceptions in the Roman Empire—there were incredibly wealthy gladiators, for example—but again, before the invention of modern capitalism, which is, give or take, maybe four hundred years old, maybe a little more, maybe six hundred years—before the invention of capitalism, the only way was conquest.
With capitalism, you have this magic of being able, just through cooperation, to have these compounding effects. The reason why, in the popular imagination, it’s still not very understood is that, when people see those CEOs—let’s say Elon Musk—who end up having fortunes, they think, “Damn, this guy must have stolen the money.” But if you look at the story of Elon Musk, no. It is… but again, we have complications because part of the clients of Elon Musk are the American government, and then we are entering the territory of cronyism.
Conor Doherty: Yeah, yeah.
Joannes Vermorel: But let’s go to something that is a cleaner example, which would be Ingvar Kamprad, founder of IKEA. He managed to turn a small seed into a bigger one and a bigger one and a bigger one, just with those compounding rates of return.
Conor Doherty: Well, this is the thing, because I already know the answer, having read it, but I do want to lay down a marker here. I think the way you have explained, quite concretely, rate of return—and you’ve touched on time—that to me is a bright, shining line that separates the perspective you’re presenting from what you typically rail against, which is the mainstream perspective.
Now, I want you to unpack why what you just said is absent in the mainstream perspective. Again, you’ve said things—just to give you a little bit of context here, because I do want quotes—
Joannes Vermorel: Yes, please. Please give me a refresher.
Conor Doherty: No, no, it’s fine. It’s just I want to give the… we can cut out my hesitation here. “Rate of return, not arbitrary non-economic metrics such as service levels, must be optimized.” And that’s from page 89 of the book.
So again, time—the time of that rate of return—is a core part of rate of return, and then we get into a periodic rate of return. But again, why is that different? Most people could say, “Well look, I work at a multibillion-dollar company. We’re doing exactly what you’re talking about, Joannes. We are maximizing the rate of our return.” Why is that not what’s happening?
Joannes Vermorel: Because that’s not what is happening.
Conor Doherty: How is that not what’s happening?
Joannes Vermorel: The main issue is that people know intuitively that making money is the point of the company.
Conor Doherty: Yes.
Joannes Vermorel: There are very few… by the way, it’s very funny, but an anecdote from my parents, Procter & Gamble—
Conor Doherty: Yes, Procter & Gamble.
Joannes Vermorel: It was a question they were asking new candidates who were applying to Procter & Gamble fifty years ago. It was part of the Procter & Gamble process not to fail applicants if they could not answer this question.
Conor Doherty: What was the question?
Joannes Vermorel: The question was, “What is the point of Procter & Gamble?” Literally, they were asking in the hiring process, “What do you think we’re doing here?” The correct answer expected from applicants was “making money.”
Most applicants, literally the quasi-totality of applicants who were fresh out of business school in France, could not answer this question. They would go into crazy explanations that had nothing to do with the matter. So the problem is that, if they had decided to fail applicants on this, they would have hired like one person a year.
So it was something so little known, so little understood by people fresh out of the very best business schools in France, that they had, as a policy at Procter & Gamble, decided not to fail candidates on this question, because they would not hire anybody if they were failing people on that. So it was funny.
You see, when people say it’s obvious, I say, “Oh, it was certainly not obvious.” It took, in fact, American companies to teach that to the French ecosystem, because the French ecosystem, again, we are back fifty years ago, was clueless about this sort of thing. So you see, the economic domination of the US wasn’t an accident. It was because there were a lot of people who understood earlier than other countries, including mine, how the game should be played. And that’s why those mega-corporations all came from the US, or actually a very large number of them came from the US, because the US had more people who understood how the game should be played.
Now, people say, “Making money, yes, nowadays, no problem. Yes, the company goal is making money, fine.” But now we have a lot of non sequitur. What do I mean? People leap from, “Oh yes, we make money, and that’s why we need to optimize service levels.” I say non sequitur. How? It does not follow. It is like saying, “We know we need to make money, and thus I do a sacrifice to appease some gods.” Again, it does not follow. You don’t have to—
Conor Doherty: I know the point you’re getting at, but I have to jump in there. Sorry. I work at Lokad, and I have to push back and present. Joannes, if you say to me service levels, which essentially represent the promise I make to my customers, whose money I want—if you’re telling me that service levels, maintaining a high level of service, does not represent an economic metric of some kind, that’s hard for—
Joannes Vermorel: Okay, unpack why that is. Don’t just say that it is. Unpack why. Service level is a percentage. It’s a percentage of availability for a given SKU.
Conor Doherty: But if it’s zero, you don’t make money. Or do you?
Joannes Vermorel: Again, let’s take a simple example to challenge this assumption, what you just told me. Apple—what is the service level of Apple right now guaranteed for the iPhone 6?
Conor Doherty: The iPhone 6?
Joannes Vermorel: Yes. What is the service level that currently Apple guarantees for the iPhone 6?
Conor Doherty: Literally, I’m going to guess zero.
Joannes Vermorel: Exactly. Zero.
Conor Doherty: Is that fifteen years ago? I have no idea. It’s completely obsolete. It’s ancient history. But that’s not Apple’s promise to its current market base.
Joannes Vermorel: So, what Apple promises is not a service level. We have just demonstrated that the iPhone 6 is at zero. So what Apple promises is a quality of service. What they promise, and what people expect, is that, if you buy this iPhone product, whatever it is, it’s going to be the very best one that has ever existed. So, the promise is: if you walk into an Apple store, the iPhone that you will get, if you’re willing to pay, will be the very best phone.
Conor Doherty: The very best phone that has ever been made on earth.
Joannes Vermorel: Yeah, obviously at scale. But you see, that is the promise they are making. It has nothing to do with this percentage availability that we just discussed. Clients, their expectations, the quality of service, are something much deeper than the percentage.
For example, if Apple tells you, “Dear customer, today I’m not going to sell you what is here. Really, I think that’s not exactly what you want. Why? Because tomorrow I’m receiving the next generation. I’m very sorry. This one, until today, was the best one, but tomorrow we have the better one. So maybe, dear client, are you okay to just wait one more day so that we can fulfill the promise of you having the best phone ever produced?” You see? And obviously people would say, “Oh yes, absolutely. That’s what I want.”
It’s an expensive thing, more than a thousand euros. Yes, I can wait one more day if you tell me that tomorrow… and maybe a billionaire entering the room says, “No, give me one today and I will buy a second one tomorrow. I don’t care.” But that’s not going to be your average customer reaction. For the vast majority, that’s why I say service level is just a very bad proxy. It’s a proxy, but the quality of this proxy is so bad that almost invariably you should not use it.
Conor Doherty: Again, I like the example. I think an example—you gave it in a previous episode, in fact—was of, let’s say, a fashion company. Do you maintain 96% service level right at the end of summer, when you’re about to phase out your lines and launch your winter coat collection?
Joannes Vermorel: Yes. Probably not, because that would be kind of mental. Like the day, I don’t know, September 30th, are you still going to have 99 or 98% service level for your swimsuit? If you do, you’re probably going to lose a lot of money the next day when October hits, et cetera.
Let me give another example where we enter even stranger territory. What would you think of a merchant that says, “I’m going to sell you this thing, but I can take it away at any point in time in the future, and you will thank me for that”? It sounds strange. I’m selling you something, but, you know what, part of the contract is that I can take it away from you, and you will agree to that, and you will even be thankful that I did that.
Conor Doherty: Like a software vendor or something?
Joannes Vermorel: No. Aircraft parts for retrofits.
Conor Doherty: That’s correct. Yes.
Joannes Vermorel: Aircraft parts—when you are an OEM—security is paramount. You do not want to kill people. I repeat: you do not want to kill people. The rule of the game is that people need to arrive alive.
Conor Doherty: On time, alive, is what we aim for.
Joannes Vermorel: Yeah, exactly. As they say in aviation, a good landing is a landing where everybody in the plane can walk away from the plane unhurt.
Conor Doherty: Yes.
Joannes Vermorel: And an awesome landing is when the plane can even fly back afterwards. But a great landing is everybody being able to walk away in one piece after landing. So this is a rule of the game. And in terms of the money machine, if we don’t respect this environment, the money machine stops, because if you kill someone, the machine just stops. You don’t need a regulation. It will just stop.
What am I talking about? If I sell you a part and then I realize afterward, “Oh crap, there might be a problem with the part”—I’m not even sure, because those people are completely paranoid—if they have even a suspicion of a suspicion that the part might be compromised, that it might endanger people, they are going to do a retrofit. So they are going to say to every single person that ever got one part from me, “You’ve got one week to send me the part back. You must. This is important. Will you sign a contract with me saying that you would?”
So now, you send me the part, and meanwhile I send you the new part. Every part that I ever sent, here is a new part. And yes, we have both agreed initially that doing things like that would be ideal. So you would have a new part. Your aircraft will keep flying with a new part that is safer. But me, I don’t want my reputation to be damaged by the fact that maybe there is a little probability that something went wrong.
And the retrofit game is: if you don’t give me every single part, I want every single part ever produced accounted for. If that’s not the case and the plane crashes, it will be on you. Not on the OEM doing the part, but on whoever failed to return the part, because you had the good part.
So you see, it can get strange. This is why, again, try to understand service level from this perspective. It’s mental. It is such a bad proxy. It is not even remotely close to approximating the game being played, which is making money by never, never killing anybody in the process. That’s aviation. You want to transport people in an incredibly safe way.
Conor Doherty: It’s certainly an imperfect proxy. I don’t think anyone would dispute that based on… we could list examples all day. What I want to push forward on is a key part of the rate of return argument, and it actually touches on the idea of time again, and it touches on your criticism of the mainstream in a concrete way.
The mainstream supply chain theory—you make the point that the idea of trying to make your decisions or even measure rate of return with time horizons like week, month, quarter, year is arbitrary, and it lacks a certain amount of precision and granularity. You capture this idea and describe it using the term “a periodic rate of return.” That’s fancy, but break down what the idea is and why a much more granular understanding of time influences the money that you make.
Joannes Vermorel: So the thing is that rate of return, as a concept used in finance, is fundamentally an aggregated concept. It’s a macro concept. It just says: imagine you are a finance guy and you just want to look at the rate of return of your investments over a long period of time. So this annualized investment kind of makes sense because you are comparing apples to apples. You know, you are comparing: if I invest in this company, what is the projected rate of return? If I invest in this company, what is the projected rate of return? Apples to apples. I can normalize everything by looking at yearly returns.
Typically, just for the audience, the rate of return that you can expect from, let’s say, the stock market is about something like 7% a year, S&P if you’re going to… and by the way, 7% means risk. If you say, “I want rate of return with zero risk, so I will never lose any money,” what you can expect is probably about 2%, if you’re lucky. If you say, “I am willing to have bad years, so years where I lose half of what I have, but I can play the long game, let’s say twenty or thirty years,” then 7% is what you can expect. If you are really pro-risk and you say, “You know what, I’m okay for my money not to go to half but to zero,” then you can get 20%. But then you have a very severe risk of actually getting to zero. That’s the risk-taker sort of perspective.
Okay, back to the initial point. In the finance world, because companies are human things, they are kind of slow. Even if you look at giants like Apple, it took decades for them to become giants. So it takes time. Even the fastest company—Google was one of the fastest growing companies ever—it still took them almost a decade to actually enter the stock market. So we are talking about slow-moving objects. Even when we say this is one of the fastest companies ever, it is relatively slow-moving. We are talking about a process that takes years. Thus, the annual rate of return is all that you need when you’re thinking like that.
Let’s get to the demand planner. But supply chain—the problem is that you are making decisions, so you are spending and earning money every minute.
Conor Doherty: There we go.
Joannes Vermorel: So the problem is—first, you’re spending and earning money every minute.
Conor Doherty: Yes.
Joannes Vermorel: So averaging everything at the end of the year is nice, but it’s not operational. It’s a way that tells you, after everything you’ve done, whether you did it right or not, but operationally it is not granular enough.
And then you have another problem: you are doing spendings and you are getting earnings, but those things are not connected. So how do you connect them, so that you can see the connection between what you earn and what you spend? The problem is that it’s very opaque, because your company is big, actually. So you have all this problem of, first, needing something that you can operate at a much lower granularity, a much finer-grained granularity. That’s the thing of the aperiodic rate of return. That’s just a formula I give. It’s in the annex, by the way.
Then you have another problem, which is: how do you attribute the value that is being created? Here we have a whole segment on value attribution, which is also discussed in this chapter, so that we can connect those things, what you spend and what you earn. This value attribution is a very thorny problem, but it’s fundamental, because otherwise you can be fooled. You may be under the impression that what you’re doing is right, while in fact you’re just wasting money left and right, and you cannot assess the counterfactual.
The counterfactual is that you can only know if what you’re doing is right if there is not an alternative way to do it much better. So even if you’re making tons of profits, maybe there is a counterfactual that would say what you’re doing is wrong because, if you had done it differently, you would have made even more profits. Right or wrong is not absolute. For your rate of return, you want to make sure, at the end of the day, that no money is left on the table, that your rate of return is as good as you can ever make it. That’s fundamentally what we’re discussing here. So again, two problems: aperiodic for granularity, and attribution, so that we can connect the inputs with the outputs.
Conor Doherty: Okay. Well, there’s a section, because this will come to the idea of visualization here, but the description that you give operationally—you use the term “operationally.” This is page 91. This is not a criticism. I’m literally going to ask you to explain this, just in case you’re like, “Where’s this going?” I’m going to read it, and then I want you to concretely link it to a supply chain situation, like if you were the aerospace MRO company we described earlier, or an FMCG—or excuse me, a fast fashion company.
“Operationally, the aperiodic rate of return slides a cursor along an option or decision’s cash flow timeline, computing at each date the rate that would prevail if the decision were closed out. Then it keeps the maximum value. The date that yields this peak becomes the option’s implicit horizon. In one number, the metric captures both the scale of the gain and the fastest pace at which the tied-up capital can compound.” And, as you said, the exact rule is spelled out in the annex. Okay. Explain what that looks like, because it’s a core concept. What does that look like to a demand planner on day one if they go, “That’s brilliant, I love that”? What does it look like? What happens? What’s different?
Joannes Vermorel: Let’s imagine a very simple situation where you are a retailer like Walmart.
Conor Doherty: Yes.
Joannes Vermorel: And you have an opportunity buy to get something that would be interesting. You get an opportunity buy for a very durable good. So that would be, for example, spoons. You have an opportunity. It’s a one-time opportunity. A Chinese supplier went bankrupt.
Conor Doherty: Yeah.
Joannes Vermorel: And now you can buy such a huge amount of stock of spoons—metal spoons—at an incredibly discounted rate.
Conor Doherty: Yes.
Joannes Vermorel: Okay. So you would say, “Oh, it’s nice. I can buy the equivalent of five years’ worth of spoons at a 98% discount.” I’m obviously making the example extreme for discussion. So, it’s an opportunity. I take it. I’m not short on cash. I will be able to sell those spoons with crazy margins. Not only will I be able to sell the spoons at a much lower price than my competitors, but I will make a lot more money in the process.
So I do exactly that, and I have this stock of spoons that is crazy high. Now I know that the problem is that my clientele, even if I’m Walmart, doesn’t have the capacity to absorb all those spoons, not in the short term. It’s just so much. I have another problem: I anticipate that those spoons will slowly get out of fashion. Okay, year one I will maybe liquidate half of those spoons. Year two I will liquidate a quarter. Again, why am I selling less and less? The answer is because fashion will have evolved, and this is not exactly the type of spoon that people want. It’s sufficiently stable, so the demand will not crash down immediately, but if I think ten years from now, I can say, yes, the taste of the market will have changed. It’s subtle things, but the demand, no matter the price, for those spoons will be about zero.
Which means I will start to earn money. Maybe in year one I already make a nice profit. Year two I also make an even bigger profit because I keep selling those things. But the total profit is so far into the future. It is super, super far into the future. It is mind-blowingly far into the future. If I were to really try to sell them to the last spoon, it might actually take twenty years.
Should I compute my rate of return taking all the money over a course of twenty years? It doesn’t make sense. I don’t need to compute my rate of return by waiting until I have sold every single spoon to the last. Why? Because maybe after just one year I have made a big profit, a big rate of return. So I have a rate of return that is very positive. Maybe after one year I will have, let’s say, compared to my initial investment, a 300% rate of return, because it’s such an opportunity after one year.
But if I let the second year pass, it’s twice as long in terms of time, and obviously I’m only selling half of it.
Conor Doherty: So you’re also occupying shelf space that could be something else.
Joannes Vermorel: Exactly. So my projected rate of return for year one might be 300%. My projected rate of return for year two might only be 150%. My projected rate of return for year five, for example, might only be 105%. If I look twenty years ahead, maybe my projected rate of return is just 1.01. So you have a 1% annualized rate of return. That’s a twenty-year ultimate perspective.
But then hold on. If you say, because you need to run this game till the end, and now you’re at 1% annual rate of return, then you say, “This is not an opportunity at all. One percent a year is worth nothing.” Hold on. You’re forgetting that in year one you had 300% return. So obviously, if you just compute naively this rate of return until the end of time, you end up with a number that is very, very low. But this is a stupid number because you could just pause your analysis at year one and say, at year one I just look at how much money I’ve made, and now I just throw the spoons away. I donate them to, let’s say, a metal foundry. I don’t care. I just donate them to whoever is willing to get the metal for free, and then I cut the time horizon. It’s arbitrary. I cut the time horizon, and that gives me a much higher rate of return.
This is obviously the correct way to think about the rate of return. I cannot say that, at one point in time, I had such a good rate of return, but then later it diminished. No, because if, at some point in time, I get my peak rate of return, I should just pause my analysis there. If, at this point in time, there is nothing to be done with what is left, I should just consider it a sunk cost. I should just throw it away or donate it to whoever is willing to get it for zero.
So the aperiodic rate of return is just a heuristic to say: you look at the entire timeline and you pick the point in time where your rate of return is the highest. This is what you mean when you have a game where earnings never stop, because the reality in supply chain is that earnings and spendings never really stop. The spending due to storage space will continue indefinitely as long as you have one unit laying on the shelf.
Conor Doherty: Yeah. Exactly. And there’s opportunity cost to that as well.
Joannes Vermorel: And the earnings can still drop in very, very slowly. So the problem is that you have this tail that is very long, but also very insignificant. It’s very insignificant.
Conor Doherty: And how is this different from… you’ve given that example. How would the service level perspective or the safety stock perspective fit into that example?
Joannes Vermorel: Safety stock and service level, the non-economic metrics, are non-economic, and more profoundly here, they are non-temporal.
Conor Doherty: Okay.
Joannes Vermorel: So, you see, time does not even exist in the other perspective. The truth is that they do perceive time, but in a very strange way. It’s not what people really think. They are embracing a stationary perspective. The stationary perspective says, “I am playing a game that has no beginning and no end.” So time—I am just repeating myself endlessly in a loop, and this loop is stationary. This is just like a sinusoid, a curve that oscillates from the beginning of time to the end of time.
So, in service level and safety stock, they embrace a stationary perspective, which is an extremely strange perception of time.
Conor Doherty: Sorry, but pragmatically, from the example that you just gave, the aperiodic rate of return would tell you basically after year one you can pretty much cut your losses here. You’ve more or less maxed out your realistic return.
Joannes Vermorel: You don’t need to answer this question yet, because maybe one year from now you will realize that those spoons are fashionable again, and your assumption about the future was actually fairly incorrect.
Conor Doherty: Okay. So it can change.
Joannes Vermorel: It can change. So the aperiodic rate of return just tells you: this investment in front of me, based on the information that I have, is it a good investment?
Conor Doherty: Agree.
Joannes Vermorel: That’s it. It just tells you, at a very granular level, should I make this allocation of resource or not?
Conor Doherty: Agree.
Joannes Vermorel: That’s it.
Conor Doherty: And what I’m saying is: to contrast this perspective with the other non-economic metrics that you criticized, how would that differ? So you are the exact same Walmart. I have a million spoons, but now I’ve got service levels.
Joannes Vermorel: Service level will tell you that you need to enforce this commitment, this percentage, and thus, to enforce this percentage, you need to allocate resources. It never asks the question of whether your resources are being allocated in a profitable way or not. It just says you must. A service level says you must allocate resources. If you say this shelf is 99% service level, it says you must allocate.
And what if you are losing money like crazy? Just to give you an example, if I say strawberries, 99% service level at the end of every day for a grocery store, I will be throwing away tons of strawberries. The service level does not care.
Conor Doherty: It will just say, by vertical—
Joannes Vermorel: It will just say, “Put the strawberries on the shelf. This is your commitment. I don’t care about profitability. Just do it.” So, you see, the service level forces you to allocate even if there is no profitability. And conversely, it does not tell you to overinvest when, in fact, it is very profitable. So, for example, back to our spoon opportunity, if Walmart is thinking, “Okay, what is my service level for spoons?” I say, why should I buy this enormous bulk amount of spoons at such an incredibly discounted price? I don’t care. My service levels are already met. I don’t need to seize this opportunity. It is irrelevant.
So, you see, it’s both ways. The service level forces you to invest when it’s stupid to invest. But conversely, even when you have an extravagantly good opportunity where it’s like, “Oh, how can I miss this opportunity?” it just says, “I pass. This is done.” This is so dumb. That’s why I say it is a proxy, but it’s so mediocre, so mediocre.
You know, it is the sort of proxy that would say, “Great artists have a name that starts with the letter L,” just because Leonardo da Vinci and Karl Lagerfeld exist. So maybe the letter L happens to be a proxy to find some great artists. But when you think of it, it is a very, very crappy proxy. You see? It is very easy to invent proxies that look good on a toy example. “L” for last name starting with L is going to be a great artist example: Leonardo and Karl Lagerfeld. But then you realize, shouldn’t it be V because da Vinci? You see?
So, fundamentally, service level is a very bad proxy that academia managed to demonstrate the value of by cherry-picking crazy examples, using crazy thinking such as the stationary perspective. Thus it is kind of semi-convincing to people, but the reality is that real-life supply chain practitioners don’t even do that. They will see opportunities when they see them, because they are not idiots.
Conor Doherty: Okay. Well, I’m going to push forward from this point, because I think we’ve dug down as far as we’re going to right now, and I do want to touch on at least one more point before we draw to a close. It’s actually related to the points that you’ve just made, or an extension of them.
You decry, perhaps with warrant, the idea of non-economic metrics, and you say they are imperfect proxies for profitability. Fine. The thing is, you do also introduce some other heuristics which some might argue are hazy or difficult to nail down, and you call them “shadow valuations.”
Joannes Vermorel: Yes.
Conor Doherty: And you argue that this does warrant some context. So, just for anyone who doesn’t know, you make the argument in the book that some effects—and I would say shrewdly you make this point—some effects don’t show up in the ledger. That’s how you put it. So again, goodwill with your clients—where does that show up? Supplier distrust, consumer fatigue. A classic example you gave before is conditioning clients to expect discounts. Where does that show up?
So you make the point that practitioners must, quote, “introduce shadow valuations.” Now, you do disclaim: these figures are assumptions because there is no historical baseline. That’s pages 125 to 126. So before we get into some of the potential criticisms or how it works in practice, expand on this concept—how it works and why it’s important.
Joannes Vermorel: It’s critically important. Making money is good because, in real life, unlike what Hollywood depicts in their movies, CEOs don’t behave like the CEO of a Hollywood movie. So, let’s start with that, because we’ve covered that, so let’s not spend too much time on this. It is important. What Hollywood depicts is ultimately people who are incredibly short-term thinkers. They will make money very quickly using the worst thing possible. They behave like Gordon Gekko.
Conor Doherty: Yes.
Joannes Vermorel: They behave like gangsters. They will just burn everything. The reality is that good CEOs—Steve Jobs and whatnot—they were looking far into the future, so far that all your measurements become irrelevant. Because the future will become what you make of it. If you are ambitious enough and smart enough and visionary enough, you can see things where you don’t have any measurement.
Just imagine: there is not a single phone on the market nowadays that is not kind of a copy of the iPhone. The original iPhone—Steve Jobs knew that. When he thought of the iPhone, he thought of the endgame, and he was right on so many things. The only thing that I would say he was not completely right about is that he thought making it relatively small—and it was already a big phone—was ideal because it’s the size of the hand. It turned out that people want to consume media, and so what has won was the iPhone, but bigger. So I would say Steve Jobs on the iPhone was like 99% correct, and there are a few technicalities, such as the ideal size of the phone, where he was wrong.
Now, this valuation—
Conor Doherty: Yes, but this is key.
Joannes Vermorel: Exactly. Where does this critical information enter your model? You have an insight that is so important. It’s not going to show up in the ledger because Apple has not even started to sell the first iPhone yet. Is it unimportant? No, it is critically important. Thus, what I’m saying is that there are numbers in dollars that cannot show up in your ledger, and they are critically important because they represent the future.
The shadow valuation is a mental trick, if you want, to have those human insights about the future embedded into your economic calculation. The problem is that, if you remove shadow valuation, what you have is just like driving a car looking only in the rearview mirror. Your transactions are your rearview mirror, and they tell you what is in the past, but you need to look forward.
And then you would say forecasting, but no, no, because again, statistical forecasting is just extrapolating your statistical data. The problem is that statistical forecasting is incredible, but it’s just a fancy way to look in your rearview mirror. It’s better, but it’s still the rearview mirror. So how do you start to look at the road ahead? The answer is shadow valuations.
Conor Doherty: Yeah. And examples. So let’s say, okay, now I am going to tip my hand a bit. For example, stockout penalty—would that qualify as a shadow valuation, in your opinion?
Joannes Vermorel: It is part of it. The problem is that… again, this is a spectrum. The problem with stockouts is that stockouts are happening all the time.
Conor Doherty: Yes.
Joannes Vermorel: So the question will be: if you are in a business where stockouts happen all the time, then the stockout penalty is something where you can look in the rearview mirror and you’re good.
Now let’s consider a business where a stockout is not every day, where, in fact, it should never happen. You are an electricity provider. What is a stockout? It’s a blackout. So, no more electricity. I was in New York during the blackout of, I think, 2006. I was in the city at the moment it happened. The last time it had happened was like thirty years ago. So blackouts in New York happen every thirty years or so.
How relevant is your historical data to support your analysis of the upcoming blackout? It is such a thing in the far past that, frankly, it’s bad historical data, first because it’s super far in the past, and just because New York itself has evolved and transformed. So what I’m saying is that, if we took something like a stockout: if you are in a business that faces stockouts all day long, you can use your rearview mirror. So stockout will not be a shadow valuation. If you are providing electricity, and this is something that should never happen, this is a shadow valuation. You see? It depends.
Conor Doherty: Well then, could you give an example that relates to the ten million people who might read this book? Because most of them will not be working for energy providers.
Joannes Vermorel: Yes. So, shadow valuation would be, for example, the client attrition that you are going to have. Any stockout will be a small thing, and the question becomes: when those stockouts accumulate, how fast do you lose your clients? This is a very nuanced question, and that’s where sometimes you have to do thought experiments, because the actual experiment will be way too expensive.
Just to give you an example, let’s go back to Walmart. It’s something that is very well known. Every parent knows it.
Conor Doherty: I was about… I know the example you want to give, and I was about to challenge you.
Joannes Vermorel: Exactly. You know the pain. You’re in this period. So: diapers.
Conor Doherty: Yes.
Joannes Vermorel: As young parents, if you don’t have diapers, you are in deep, deep problems. So you want diapers, but you don’t want any diapers. You want the brand that you like, that your baby likes, because when you change diapers, it’s a pain. It’s a pain. It’s a real pain. Thus you are very, very demanding on that.
Now, because it is common sense, are retailers willing to experiment with stockouts on diapers? No, they’re not. They’re not. So diapers are always, always super nicely stocked precisely because of that. So we have almost no data on stockout of diapers, because retailers who know that it’s so critical never let themselves run out. And when I say never, it’s so rarely. I suspect Walmart has probably above 99.9% service level on diapers. So it never happens.
Thus, the question becomes the counterfactual, which is… then the shadow valuation—that’s where I was going with the original question. We almost never observe stockout because they’ve added value to it. But it is a thought experiment, because we know that stockout on diapers will cascade into losing parents, and this and that and that.
Conor Doherty: Not only that, it’s losing parents, but it’s also losing the other things that would have been bought alongside, because, as a young parent, I don’t just buy the diapers. I also buy the formula, and I want to buy groceries. I’m not going to multiple stores. So I go in. The first thing I look for is diapers. If there are none, I go to the next available store.
Joannes Vermorel: Yes.
Conor Doherty: So this is why I gave the example of stockouts and stockout penalty, because I know—and again, anyone who’s read the book can intuit and apply it—that there are certain things where, if you have a stockout, your losses are not just limited to that thing. It is everything that would have been bought in combination, or in the basket perspective, I think you use.
Joannes Vermorel: And as a young parent, if you go twice in a row to a store and twice—
Conor Doherty: Yeah, yeah. I’m not going back. I’m not going back.
Joannes Vermorel: Then you’re not ever coming back. So it’s not just the basket; then you have lost the customer forever.
Conor Doherty: Exactly.
Joannes Vermorel: Again, retailers are smart. People at Walmart are smart, so they know that. So they don’t play the game. Thus they never have any data, almost, because by definition they don’t want to play this game. But because they need to have an economic reasoning, at the end of the day you still need to decide how many diapers you have in stock. Ultimately, you still need to make an investment, even if you know that your historical measurements will not be an accurate measurement. The justification is the shadow valuation. It is a way to rationalize.
And then the interesting thing—why do you need those shadow valuations? Because, in fact, you are going to have many shadow valuations. Now it’s going to be: okay, we need to have that. But guess what? For example, Walmart, I think now in the US—again, I don’t live in the US—I believe they’re selling insulin too.
Conor Doherty: I don’t know.
Joannes Vermorel: So insulin is critically important, and that’s the same thing now. So that means that my shadow valuation for diapers, which is critically important, is going to compete with my shadow valuation for insulin.
Conor Doherty: Because you’re allocating resources one way or the other.
Joannes Vermorel: Exactly. And now I have a game where I will be able to, as a thought experiment, balance those things so that I am consistent. Again, the problem—and that’s where we are back to economics—is scarce resources. At the end of the day, even a company that is as fantastically rich and powerful as Walmart has only a finite amount of resources. Thus, those resources need to be wisely spent. That’s where rate of return comes in, and the shadow valuation is just to make sure that you’re playing the long game, so that you can make a lot of money because you’re looking far into the future.
Conor Doherty: There’s another anecdote to demonstrate the point. I don’t think it’s listed explicitly in the book, but again, I work at Lokad, so yeah. It’s one I like because we are actually talking about it. You gave the example of discounts, and discounts can instill a negative or financially injurious behavior—counterintuitively—in customers.
Anecdotally, I have seen that a couple of times, and I know people who really like certain coffee brands at Monoprix—it’s a big chain, basically the French Walmart, but not as big, but as common. Anyway, there’s a certain brand of coffee and he only buys it when it’s on discount. Literally, he’s like, “When it’s on discount, I will buy everything I see.” When it’s 50% off. But when it’s full price, I buy anything other than it, because it’s not worth it at eight euros, but it’s worth it for four.
Joannes Vermorel: And here you’re behaving exactly like the fictitious buyer I was discussing with the Chinese spoons. You say, “It’s an opportunity.” Yeah, I seize the opportunity. If you were thinking in terms of service level, this is dumb, because in fact you’re going to be at 100% service level for months. You go every week to the store and you’re buying coffee for three months. How does that make any sense from a service level perspective? You are, by definition, at 100% for way beyond your lead time.
What you’re doing is something that is unthinkable from the service level perspective. But as soon as you embrace an economic perspective, it makes perfect sense. You have an opportunity, you seize it. Then the problem becomes, for you, in your kitchen—
Conor Doherty: Yeah.
Joannes Vermorel: Am I going to take six months’ worth of coffee, and that’s going to take all my shelves? Because you’re in Paris. Apartments are expensive and whatnot. So probably what drives your limits is, first, I’m not going to walk back home with twenty kilograms of coffee. That’s probably constraint number one. And then the second constraint is, my wife is going to be mad if I take up three entire shelves of coffee bags.
So those are your limiting factors, but they have nothing to do with the service level. Again, we go back to service level being such a terrible proxy.
Conor Doherty: Well, we’ve been going for quite a while.
Joannes Vermorel: Yes.
Conor Doherty: So I am going to draw things to a bit of a close now. But I will point out that we’re now, I think, 130 or 140 pages into the book, four chapters. A lot’s been covered. We’ve spoken now for well over four, probably five hours. Where do we sit in this playbook now? For people who’ve made it this far, what do they have to go forward with now? If people stop reading now, what do they have?
Joannes Vermorel: They have an understanding of how they can actually start trying to make their company earn more money. We have now started. We have now, you know—
Conor Doherty: Yes, yes. We have already arrived.
Joannes Vermorel: Yes. Now we don’t just understand what supply chain is. We understand how it connects to making money. Thus, I would say, if we stop and there is nothing else in the book, people can start reinventing the rest on their own, and they are properly equipped to actually reach the point of profitability, because now the game is on. You play with that, and you’re going to invent stuff that will make money.
So, you see, we have already reached a point where, now, if you understand truly those four chapters and you invent a new supply chain technique, first, you are inventing something that belongs to supply chain—that was chapter three—but now you will invent something that you will be able to assess as a net positive contribution, value-wise, to your company.
So, in the previous chapter, we clarified whether you were even inventing something that belongs to supply chain. Here, we are saying: now, are you inventing something where you have a criterion to say, “Am I on the right track?” And here you have the criterion to say if what you’ve invented is on the right track. That’s making money. That’s what we have.
But then I will tell the audience that the rest of the book will actually start getting into the nitty-gritty details of what is there to be invented to do that. But here, you are prepared now, if the audience wants to do that, to actually reinvent the rest of supply chain. But they will be reinventing the right stuff.
Conor Doherty: So greed is good.
Joannes Vermorel: Okay, good. Greed without coercion. That’s the thing. It is such a nuance, and by the way, it took almost a millennium and a half for Christianity to kind of get that. I said it as a joke, but no, it’s important. You have to think that usury, because rate of return is very deeply related to usury—usury was considered a sin by the Catholic Church. I’m personally Catholic, so I know the history of my own religion. It was considered a sin.
But guess what? What is the country, what is the heart of Catholicism? The answer is Italy. Italy only emerged in the nineteenth century, so I’m talking about the geographical region, not the political entity that only arose very late.
Conor Doherty: Napoleon designed the flag. Did you know that?
Joannes Vermorel: Yes, yeah. But the thing is, accounting was actually—I think double-entry accounting was only invented in the fourteenth century by an Italian monk. So, you see, that’s the grand paradox. Usury, rate of return, was seen as a sin, but who invented the tools to actually maximize rate of return? That’s double-entry accounting. This is the incredible invention. With double-entry accounting, we have the elements to start playing the game of capitalism right. Without double-entry accounting, nobody knows how to even count the score. You don’t know if you’re losing or earning money.
This is why I say it is such a landmark invention, because it is the start where you can play the capitalism game and you know the score. You have this mechanism. It’s so fundamental. And guess what? The very first banks are going to emerge in Italy, and the very first people who are going to play this non-aggressive sort of new rule of making money… that’s the thing. The beauty of it is that it’s about creating almost value out of thin air.
That’s why, by the way, there was a quote from, I think, Einstein, saying, “I am baffled by the fact that compounding interest might actually be the greatest force of the universe.” It was this sort of bafflement of contemplating that it seems to be creating—to a physicist, where nothing is created, everything is transformed, that is the motto of physics—it seems, to a physicist, that when you look at those compound interests, you are creating stuff out of nothing. And it’s a little bit like that. That’s the beauty of it.
But it is so strange that even nowadays—and that’s what I’m describing in this book—the mainstream perspective does not see this game. They do not understand this game, creating value out of nothing. If I give concluding thoughts, the way they see supply chain is not as a forever-compounding game. They see it as cost minimization. You have a goal—that’s your service level target, that’s your commitment, whatever—so you have your top-level constraints that are laid down, and then you minimize the cost to get there. That is the mainstream supply chain perspective. It is a cost-minimization perspective. It is not a rate-of-return maximization perspective.
Conor Doherty: All right. I’m sorry I quoted Wall Street right at the end, because it seems to set you off. But again, a nice concluding thought.
Joannes, we’ve been talking for quite a while now. I don’t have any other questions. Thank you very much for joining me in the Lokad studio, and I look forward to discussing chapter five. And to everyone else, thank you for watching. If you want to continue the conversation, you can always reach out to Joannes and me on LinkedIn, or send us an email at contact@lokad.com. And with that, we’ll see you for the next episode on chapter five. And yeah, get back to work.
