What We’re Reading (Week Ending 08 September 2024) - 08 Sep 2024
Reading helps us learn about the world and it is a really important aspect of investing. The late Charlie Munger even went so far as to say that “I don’t think you can get to be a really good investor over a broad range without doing a massive amount of reading.” We (the co-founders of Compounder Fund) read widely across a range of topics, including investing, business, technology, and the world in general. We want to regularly share the best articles we’ve come across recently. Here they are (for the week ending 08 September 2024):
1. #360 Robert Kierlin: Founder of Fastenal – David Senra
What’s the most important part of Fastenal’s success that outsiders discovering the company for the first time don’t understand? The number one thing is the people aspect. The goal is to unleash entrepreneurial passion, a commitment that I will be self-driven to do better than what you can expect. It is a mindset.
This is what they’re telling their employees, “Run your business like you own it.” When you trust people to solve problems and make decisions and you let them go, that’s where the magic happens. That is the story of this company. Fastenal embraces a spirit of radical decentralization in autonomy.
“Each of its 2,700 stores operates as a stand-alone business with a clear leader and full P&L responsibility. We grow from the ground up based on the actions and decisions of thousands of people who run their businesses like they own it. I want those people to stay with us forever. They will never have to stop at a certain level.”…
…“We are now one giant organization. We are 2,700 small businesses wrapped up into one big company. Society tells us, you’re a big company, act like it. We say no, you don’t get to define us, we define ourselves. We’d go against the grain in almost everything that we do.” And so there’s just crazy stats about the company that, “More than 95% of our current batch of general managers have been promoted from within”.
“For nearly all of our senior leaders have worked their way up from entry-level positions.” So that leads into the second thing I want to tell you about, which is this interview with the CEO that had succeeded Kierlin or Bob — we call him Bob, when he stepped down. His name is Will Oberton.
Oberton started out at Fastenal as a store clerk. So he went all the way from — they mean business. He went all the way from store clerk to CEO…
…“By keeping operating costs very low, Fastenal is able to pay their employees incrementally higher wages, and thus, more effectively develop and retain talented salespeople. The quality of service and depth of knowledge that the employees have eventually brings in more revenue, which grows the business and allows it to further lower operating expenses as a percentage of revenue, thus allowing for more hiring of top-quality employees, which brings in more revenue. This is an overlooked virtuous circle of sorts.”…
…Why is it so important to have everybody working as a single cohesive team, to have everybody thinking that their role that they’re playing is just as important as the person next to them? “At Fastenal, we believe that you could be the best salesperson in the world. But if the order-picker doesn’t pick it right or the truck driver doesn’t get it there on time or the billing clerk doesn’t bill it correctly, you end up with an unhappy customer.”
“Everyone is key. You are better off working to make everyone equal so they stay focused” and he goes back, what do you think he’s about to say? I bet you can already finish his sentence for him. “You are better off working to make everybody equal so they stay focused on the common goal of pleasing the customer.”
[00:26:01] He’s going to give us more advice on how to do that. You need to install a reward system that keeps everyone focused on the common goal. He’s talking about incentives. If you have Fastenal’s common goal of growing our company through customer service, you will avoid any rewards that don’t fit that goal. And so when I got to this part of the book, I thought about Charlie Munger’s like three rules for incentives.
And so this is what he said, “Number one, everyone underestimates the power of incentives. Number two, never ever think about anything else before thinking about the power of incentives. And number three, which Bob is nailing, the most important rule in management, get the incentives right.”
And again, you have to be careful of these subgroups that are going to naturally develop in your company because his whole point is like “Listen, your incentives have to — they have to fit your overall common goal,” right, the common goal of pleasing the customer. And so he gives us an example, “If you do these incentives based on like separate groups, they can optimize for things that go against your common goal.”
So he gives an example that this is a really smart idea. “We do not reward production people for minimizing scrap. If some of that scrap you eliminate comes from the extra parts that guarantee you have a full order quantity ready when the customer wants it.” The incentive superpower that Munger talks about, you clearly see by picking up the book…
… I want to go back to that story of the CEO that was meeting with Buffett, the CEO that succeeded Kierlin. So this giant part of Fastenal’s business now after, this was invented after this book was written, okay, the first version in 1997, was the fact that they have these vending machines.
And the way I think about the vending machine is like think of anytime you’ve been in like a hardware store, right? You’ve got ACE Hardware or Home Depot or anywhere else. And think about how all the equipment and supplies are presented, kind of like searching through, it’s kind of like a chaotic mess.
So Will Oberton, which was the former CEO, but he’s no longer CEO now, but he’s the one that was CEO after Bob, okay? Oberton also developed an industrial vending machine system. There’s a video on YouTube that’s fascinating about this. It’s from Fastenal. Fastenal has their own YouTube channel. You can see the vending machine if you just type in Fastenal vending machine, if you’re interested in this, I thought it’s actually cool.
Oberton developed an industrial vending machine, and I searched for it after I read this because like I got to see what this looks like. Oberton had developed an industrial vending machine system, helping Bob realize a lifelong goal. In 1951, as a 12-year-old working in his father’s auto parts store, Bob was bothered by the fact that his dad had to send customers searching for nuts and bolts to someone else’s store.
He imagined that a vending machine installed at his father’s place might pop out fasteners like gumballs. Once on his own, he tried to convert a cigarette vending machine to this purpose. He couldn’t get it to work. So he started selling fasteners over the counter. Thus, Fastenal was born. 40 years later, working with a snack machine manufacturer and off-the-shelf software, Will Oberton got the job done.
Fastenal’s vending machines have been a big hit with customers. So their vending machines are actually installed in their customers’ locations. It cannot get simpler for this. You got to watch the video, I’m telling you. Oberton got the job done. Fastenal’s vending machines have been a hit with customers, generally helping them save 30% on supplies.
[00:46:05] The machines have cut down on theft and enabled automated reordering. That 4-year-old business, which I think now is like 15-years old, within 4-years old, this new idea already started contributing to 36% of the overall sales of Fastenal, I think it’s like over 40% now.
2. A French Bank Like No Other in Europe Seeks to Export Its Model – Phil Serafino and Albertina Torsoli
Bpifrance is a bank like no other in Europe.
The French lender has made more than €50 billion ($56 billion) in loans to small and mid-sized businesses and has €52 billion in stakes in almost 1,000 companies. It has backed everything from a startup wanting to take tourists to the edge of space in balloons and a chain of trendy Parisian nightspots to the automotive giant Stellantis NV. A force to reckon with on French deals for M&A advisers like Goldman Sachs Group Inc. and JPMorgan Chase & Co., it has lured away bankers from firms like UBS Group AG and Rothschild & Co…
…No other European country has an agency quite like Bpifrance: a for-profit, state-owned merchant bank with a mandate to foster national champions. Its wide-ranging lending activities are financed largely by borrowings guaranteed by its ultimate backer: the French taxpayer. And for all the political turmoil at the moment in France, its interventionist policies are likely to find favor no matter which coalition — from the left or the right — ends up forming a new government.
More than a decade after it was created under then-President François Hollande and his economic adviser — one Emmanuel Macron — Bpifrance exemplifies 21st-century French capitalism: Entrepreneurs build businesses with cash, nudges and nurturing from the state, which in turn wants them to create jobs at home and develop innovative technologies. Explicit in the deal: The government will fend off foreign interlopers if necessary…
…Bpifrance’s investment prowess and risk management haven’t really been tested because Dufourcq hasn’t faced a prolonged economic downturn, enjoying a favorable wind at his back almost from the start — even during the pandemic, when the French state opened the cash taps to prevent businesses from going under.
Its stock-picking bets also haven’t always paid off. A stake in train-car maker Alstom SA, for example, has lost about a third of its value since the investment early last year. Shares of Stellantis, in which the bank has a 6.4% holding, have slumped about 45% from their peak in March as the carmaker struggles to fix problems at its US and European operations.
Also, for much of the bank’s existence, it could finance itself at rock-bottom interest rates, something that’s no longer the case. A slowing economy and higher rates also may start to hurt companies that borrow from the bank: Bpifrance’s loans classified as doubtful stood at 4.7% at the end of 2022, up from less than 4% in recent years, according to the bank’s annual reports. It didn’t disclose the statistic in its 2023 report.
Dufourcq shrugs off such concerns, noting that the three decades-old agencies that combined to form Bpifrance survived some deep financial crises, and says his bank often says no to risky investment proposals.
While some European countries have national development banks, Bpifrance is unusual for the breadth of its offerings. It operates 50 offices around France, often sending representatives door to door to drum up business. In addition to debt and equity investments, it offers financing and credit insurance to exporters and training and consulting services to entrepreneurs — including on how to shrink their carbon footprint.
3. How Richmond Fed President Tom Barkin Sees The Economy Right Now – Cale Brooks, Tracy Alloway, Joe Weisenthal, and Tom Barkin
I talked to someone from Germany yesterday — this is going to make me interested but not your [audience]. Our savings rate went up at the beginning of Covid to about 15 or 16%. Same thing happened in Germany. Our savings rate has come down to about three and a half. Theirs is still at 17. So, why are German consumers not spending the way American consumers are? That’s an interesting topic. It’s something we spent some time on. It’s the kind of thing we spent some time on…
Tracy (08:01):
What’s the theory?
Tom (08:04):
Well, so the thing that really makes it crazy interesting is, there’s a whole social safety net in Europe that doesn’t exist here. And so, most of the time you think people are saving for retirement, they’re saving for a rainy day, or they’re saving because they’re worried about losing their job. Well, in Germany, they kept everyone’s job during the pandemic and you’ve got a pension. So why are they saving? And I think the best explanation I’ve gotten, it’s actually something on my list to study going forward, is there’s just a lot more precautionary feeling about the situation in Europe, the risk versus the Ukraine, and what’s happening over there. And it’s just a culture that maybe has just gotten a lot more cautious due to geopolitics if nothing else.
Joe (08:43):
That does sound really interesting. By and large, I mean obviously, the situation in the Argentina economy is radically different than it is here in the US. Germany is probably still, all things considered, similar cyclically to the us. Does it feel like, by and large, at least among developed countries’ central bankers, that there is a strong set of common mysteries perhaps? Or are they really like, everyone’s sort of seeing different things in their own country? I mean I’m sure it’s a mix of both, but how much of a global factor is there?
Tom (09:16):
Much more in common than different. The whole practice of central banking has been, I’d say, globalized over the years. And central bankers really do think about inflation targeting, for example, in the same ways. And there are banks, like New Zealand and Australia, that, back in 2000 or even before that, set inflation targets before the rest of us and we learned from them. And so there’s a lot of learning, there’s a lot of discussion. I think there’s very much a common framework. Now, the economies are very different.
I mean, the US economy has come through this unbelievably well, the European economies have not. And so we have a much stronger economy. So much of our economy is services, so much is supplied to ourselves. A lot of this deglobalization is felt much more on the European side. The challenges in China right now are felt much more on the European side. And then emerging market countries, they really just are worried we’re going to increase rates further and they’re going to end up offside. And so, they’re very dependent on our strength of our dollar and the weakness of our dollar…
…Tom (11:43):
I think the economy, since we were together three or four months ago, the economy’s moved in a very different way. First of all, on the inflation side, I might’ve even said four or five months ago I was looking for inflation to sustain and broaden. So, it’s sustained. We’ve got very low readings for four months in a row. And it’s now across the basket, whereas six months ago [or] eight months ago, it was really just in goods. And so the concern about inflation, reaccelerating has definitely come down significantly. At the same time, the labor market stats have also softened. And so, the phrase I’ve been using is, ‘people aren’t hiring but they’re not firing,’ and that’s just not a high likely sustainable outcome. Either demand will continue and people will start hiring again or you’ll start to see layoffs. And so I think there’s more concern on the labor market and less concern on inflation relatively…
…Tom (13:00):
So consumers, you hear a lot of talk about people saying that consumers [are] weak and people are running out of savings. That’s not what I’m hearing. What I’m hearing is consumers are still spending but they’re choosing. And, the way I think about it is, they now have the time when they go into a store and they see something that’s at a price they don’t like to say, ‘I think I’m going to do something else.’ And so if you look at Walmart’s results, they would talk about people trading down. If you look at Target’s results, they talked about the kind of reaction they’re getting to lower prices. McDonald’s results in the $5 value meal. I’ve talked to hotel chains that every room is booked, but they can’t raise price at all because the second they raise price, people just won’t buy it and won’t book it. I talked to a fast food leader who’s rolling out software actually to encourage their franchisees not to raise prices anymore…
…Tracy (15:15):
What’s the urgency, then, on supporting the labor market? And there’s obviously a debate going on right now about how fast deterioration in that market actually happens. We had Claudia Sahm on the podcast recently and she was talking about, ‘maybe it’s different this time,’ but how are you thinking about the pace or the rate of change in the labor market?
Tom (15:37):
So the other thing that’s happening in the labor market is a lot more supply of labor, and part of that is participation, prime age participation hitting 2025 year highs, and immigration, which is up significantly. And so the last jobs report where unemployment went up from 4.1 to 4.3, you actually added jobs, 114,000 jobs. We just added 420,000 people to the workforce. So the denominator got bigger. And so, you know, there’s some people who look at the unemployment rate and say, ‘Oh my gosh, the labor market’s about to fall off a cliff.’ That’s not how I see it. I see a loosening labor market being driven by a lot more supply. Now, what’s the urgency? We’re not in a situation, I don’t believe, where there is this big cliff there, but when we make policy, you’re trying to make it for a year from now, right? Because [of] the lags of monetary policy, you’re trying to meet a year from now.
And so you’ve got a labor market which is slowly cooled and you’ve got inflation which is now gradually cooled. And so, you sort of say, ‘Well, which do I worry most more about?’ And it’s been very clear for the last two and a half years that all you worry about is inflation. And now those are much more balanced…
…Tom (21:57):
Well, I see inflation upside risk in two places. First is, we’re at 2.5% for the last 12 months. Our target’s 2%. So while we’re doing great at bringing it down from when it was once 7.1%, core is still at 2.5%. And even the most optimistic forecast for the back half of this year don’t believe it’ll get to 2% because the numbers were so good on a 12 month basis…
Joe (22:20):
We’re talking we’re about on a year to year basis as opposed to like a three month though sequential, yeah okay.
Tom (22:22):
On a 12 month basis. Because the last half of last year was also very good. And so, we’re at least six months away, even with really good inflation data, from the inflation numbers hitting 2%. And if the numbers are just pretty good, not really good, there’s a risk that we plateau at some level over 2%. That’s one risk. The other risk is I do see medium term inflation pressures that are out there. We have a conflict in the Middle East that could spiral. Deglobalization is a very real risk and that means that the imports of goods could be more expensive going forward, or if we even reshore, more expensive. Housing’s a place where, if rates artists start coming down, one of the things I worry about is that will spool up demand for people who’ve been waiting to buy a house till mortgage rates come down, but there won’t be any new houses built. I mean that effect is two years, three years out.
And so what happens if you have more demand for houses with the same kind of supply? Or even if more houses come on the market, everyone who puts their house in the market is a buyer and a seller. So you’d still have this excess of demand over supply. So those things are potential inflationary risks. Now, good policy works against that, and if we do the right thing with rates we’ll work against, but that’s why I just want to make sure I understand it and see it before I declare victory.
Tracy (23:37):
What’s been the most surprising thing that you’ve heard at Jackson Hole this year? You talked about German savings rate, but beyond that, is there anything that caught your eye or your interest?
Tom (23:48):
Alan Blinder asked a question today that I thought was pretty interesting. He said, ‘When you think about monetary policy lags, why aren’t you talking about how to shorten them?’ And I’ve said, almost as it’s a given, that when we raise or lower rates, it takes 12 to 18 months for the full effect to go into the economy. Well, part of that is because the economy doesn’t behave in a way that would allow it to happen quicker. An example: I think the number is, in 2009, 60% of [the] mortgages in this country were adjustable rate. Today it’s 8%. And so when we raise or lower rates, it doesn’t flow through to mortgages quickly and certainly not even like it did 15 years ago. And I’m not saying we should change the mortgage market, but it does make you stop and think, how much of our policy, the effectiveness of our policy tools, is a given or how much could actually change over time as the economy changed?
4. No Priors Ep. 78 | With AWS CEO Matt Garman (Transcript here) – Sarah Guo, Elad Gil, and Matt Garman
…Now that we’re at a $100 billion run rate, I think 85% of workloads are still running on-prem today by most estimations, somewhere in that range. Pick your number, whether it’s 80 to 90, or whatever it is. That’s enormous. If there’s still 10x growth of just existing workloads – forget all the new genAI workloads that are being created every day – these are just existing workloads to move, there’s a 10x number in there, so that business is massive…
…Gil (12:40): You mentioned that 80% of workloads still haven’t migrated over. What do you think are the main blockers to that today? is it just momentum? Are there specific features? Are there big things still to build?
Garman (12: 48): There’s some technologies that I think… Look, if I had an easy button, and by the way we’re trying to build an easy button, but if I had an easy button that would just migrate mainframes to a modern cloud architecture today, almost everyone will push that button. But it doesn’t quite exist today and it’s not as simple as like, “Great, I’ll go run your mainframe in the cloud.” That’s not what customers want. They want to actually modernize those workloads and have them into microservices and containerized workloads and other things like that. So that’s one, is there’s just a bunch of workloads like that that are old and and their customer’s running a big SAP thing and they want to move it to the cloud but it just takes time because it’s tied to a bunch of other things like that. There’s also a bunch of workloads that as you get out of core IT workloads that are in line of business, that are the next set of things. Whether that’s say telco workloads that are running the 5G infrastructure around the world, we’ve slowly been moving those to the cloud and helping those customers get that flexibility and that agility of of running those in the cloud as well. But they’re slower to move.
If you think about all the compute that runs factories out there today on factory floors, most of those have not been modernized. And there’s a huge opportunity, by the way for AI, to totally revolutionize how you think about factory workflows and efficiency there. But a lot of that hasn’t moved. There’s on-prem infrastructure that people are still amortising, there’s still people whose jobs it is to run on-prem data centers, and so they’re resistant to moving things. There’s a bunch of factors in there and so some of it is just takes time, some of it is technology pieces, some of that is we still have stuff to go build and innovate and help make it easier for customers to do that.
Guo (14:37): I’d love to hear about just the initial investigation of generative AI as a technology change and how AWS began to react to it, invest in it, because to some degree it puts us all back in the on-prem co-lo era of the world, where to get one of these, if you’re doing any sort of real pre-training, to get your startup off the ground, you’re back to, “I’ll buy a bunch of DGX boxes somewhere and I need to think about the cost and management of that.”
Garman (15:07): Actually most people are still buying those but in the cloud. But it’s not a serverless type of thing. Most people are still not buying H100s and hosting them in a co-lo or anything like that. And increasingly, I think that’s going to get harder and harder as you move to liquid cooling and larger clusters. It is a super interesting space. I think we’ve been working on this space for how many years now – we’ve been investing in AI broadly for the last 10 years, and it’s why we started five or six years ago investing at the infrastructure layer and building our own processors, because we knew this was coming, we saw this path coming and we knew that that’s also not a short-term investment. It’s one of those things you got to invest way ahead. And then we were investing and building generative AI models, and then OpenAI made a generational leap forward with what they were able to do, what was possible, and many people have talked about this. But it really in some ways was a discovery as much as anything about just what was possible and unleashed the new set of capabilities.
So we actually as a business took half a step back and said, “These are going to be transformational abilities and assuming that this technology gets better and better and better over time, how do we make it so that every company out there can go build using those technologies?” Different than, “How can I go build a consumer application that people are going to be interested in?”, we took it from the point of view of AWS, “Just what are the building blocks that I can help all of our customers, whether they’re startups, whether they’re enterprises etc, go build interesting generative AI applications.” We started from first principles. Customers are going to care a ton about security. That’s not going to change. They’re not going to all of a sudden not care about securing their infrastructure.
We also had two more hypotheses. One, the idea that there wasn’t just going to be one model. We thought that there was going to be a lot of models for a lot of different purposes, and there’d be big models and small models, and people would want to combine them in new and interesting ways. I think the last two years have probably played that out. I think when OpenAI first launched, it wasn’t as obvious, but that was one of the bets that we made. The third one is that we view that every enterprise that was building on us, the interesting IP that they were going to bring to the table was mostly going to be their data, and they were going to care that their data didn’t leak back into a model or escape from their environment. So we built a bunch of what we did starting from those principles of how do we make sure that these things are secure, that their data is secure, that they can have access to every piece of technology that the customers need to go build interesting applications, and they can do it in a cost effective way. That’s how we approach the space.
I think we now have a platform in Bedrock, in Trainium chips and Inferentia chips, and then a bunch of the other capabilities around as well as the suite of models that we offer, both proprietary as well as open source ones – or open weights ones. I think we’re starting to see that momentum pick up and we’re seeing more and more customers really like that story. They like that platform to build from, and we’re seeing enterprises really lean in and want to build in that space because it gives them a lot of that control that they want as they go and build applications…
…Gil (26:25): The other place that a lot of people are spending time right now in terms of bottlenecks to utilization or usage or future-proofing, is actually more on the chip side or semiconductor or system side and in terms of DC capacity. Obviously you all have been building Trainium chips and other things which I think is really exciting to see that evolution. How do you think about future GPU shortages? Does that go away, when? I’m sort of curious about how you think about forward-looking capacity, and is the industry actually ready in terms of building out data centers, building out semiconductors, all the rest of it, packaging.
Garman (26:56): I think we’re probably going to be in a constrained world for the next little bit of time. Some of these things, they take time. Look how long it takes to build a semiconductor fab. It’s not a short lead time and that’s several years and TSMC is running fast to try to ramp up capacity, but it’s not just them. It’s the memory providers and frankly data centers that we’re building. There’s a lot of pieces in that value chain that I think as you look at the demand for AI which has been – exponential might be undershooting it – some of those components that support that I think are catching up and I think AWS is well positioned to try to do that better than others are.
We’ve spent a long time thinking about – in the last 18 years, learning how do we think about smart investing, how do we think about capital allocation. We’ve spent a bunch of time thinking about how do we acquire our own power, how do we ensure that it’s green and carbon neutral power, all super important things. We’re the largest purchaser of renewable energy over the last… new contracts, so actually going out and adding and supporting new renewable energy projects. We’re the largest provider I think, each of the last four or five years. So we’ve been leaning into that for a while to ramp up this and this is just a step up. So we’re thinking about how are we acquiring enough power. Our own chips is a way to support the growth of Nvidia chips, and so I think the more diversity there, the better off we are. We’re a huge partner of Nvidia’s. Nvidia actually runs their AI training clusters in AWS because we actually have the most stable infrastructure of anyone else, so they actually get the best performance from us. We love that partnership and we have a great and growing relationship with them. We think things like Trainum are a good diversification and I think there will be some workloads that run better on Trainium and are cheaper on Trainium over time, and as well as Inferentia.
I think inference is one of those workloads that – today it’s 50/50 maybe of training and inference. But in order for the math to work out, inference workloads have to dominate, otherwise all this investment in these big models isn’t really going to pay off, so hopefully for the industry that all happens. But I think we’re probably going to be tight for the next little bit of time, because the demand is almost infinite. I mean it seems infinite right now.
5. Timing the Stock Market Using Valuations – Ben Carlson
I’ve never found a legitimate way to utilize valuations to determine entry or exit points in the stock market. Maybe when things get to extremes but even then valuations can be unreliable.
In early 2017, I wrote a piece for Bloomberg about stock market valuations:…
...This was the lede:
Something happened in the stock market this week that has only occurred twice since 1871: Robert Shiller’s favorite valuation method for the S&P 500, the cyclically adjusted price-to-earnings ratio, reached 30. So, is it time to worry?
The only other times in history when the CAPE ratio reached 30 were in 1929 and 2000, right before massive market crashes. So it made sense that some investors were worried about the stock market being overvalued.
The S&P 500 is up nearly 170% since then, good enough for annual gains of roughly 14% per year.
Sometimes valuations matter, but other times, the market doesn’t care about your price-to-earnings ratios.
The same is true during bear markets. Sometimes stocks get downright cheap but not all the time…
…Three of the four bear markets this century didn’t see the CAPE ratio come close to previous bear market valuation levels. If your plan was to get more aggressive when the market got cheap enough, you would still be waiting.
The problem with using valuations as a timing indicator is that even if they do work on average, missing out on just one bull market can be devastating. You could be waiting a mighty long time to get back into the stock market and miss out on big gains in the meantime.
Disclaimer: None of the information or analysis presented is intended to form the basis for any offer or recommendation. We currently have a vested interest in Amazon (parent of AWS). Holdings are subject to change at any time.
