[...]

It’s pretty clear what drove the market this month (and this quarter). If looking at factors, it’s beta and size. If looking at industries (which is probably the stronger contributor this quarter), it’s technology, autos, and industrials (and, thanks to Mr. Buffett, trading houses which has been the strongest performer). So one might ask us, “you’ve been seeing that move all year;  why don’t you buy more tech or machinery?” If I had the luxury of trading stocks, I would. If I could simply buy a stock because it’s going up, I could. But our philosophy is based on fundamentals. And, yet again, during the month of June when we had the most company visits this quarter, we continue to hear what we’ve been hearing all year … that there is very little to be excited about. Ironically, we were actively looking at the tech and machinery sectors toward late last year to early this year. There are a vast number of quality companies in these sectors and we’ve been anxious to add when the opportunity set was attractive. However, the weak and/or uncertain fundamentals suggested then, as well as now, that the value proposition wasn’t there. And we haven’t been going out looking for weak datapoints; that is what we continued to find throughout the year and, if anything, it has, on the margin, worsened.

​

​We’ve met with 2 large semiconductor production equipment companies (both leaders in differing fields) and they see no pickup in demand. In fact, one, who counts NVIDIA as a client, was surprised by their recent announcements. Perhaps the IR team is the last to know. Or maybe the company’s visibility is low. In either case, they are definitely surprised about how their stock is reacting and are worried that they won’t be able to deliver satisfactory results. A separate SPE parts manufacturer (i.e one who sells into SPE production companies) recently revised down their forward 6-month forecast after what was thought to be already highly conservative estimates. We’ve heard from wafer manufacturers directly and indirectly both within Japan and outside, as well as chemical companies who supply to these wafer manufacturers, that wafer production has started to slip this quarter and have begun lowering expectations for future quarters. We’ve heard from other chemical companies in the semiconductor food chain, who had thought earlier in the year that the Jun Q would be the floor and a turnaround would begin from the Sep Q, that they are now pushing out that view by least another quarter. Prices may be stabilizing as inventories have fallen, but we continue to hear weak demand. Now, I wouldn’t say that technology is collapsing. It’s just that it’s not looking any better than it did 6 months (and 40%) ago and, if anything, feels worse. A major leading electronics components manufacturer who feeds into the entire supply chain and has fairly good visibility said to us, outside of the new smartphone launches for the fall, other smartphones (notably China), PCs, and even autos are, on the margin, picking up slower than they had initially forecast at the start of the year. This is being corroborated by industry data. While the data is lagging, domestic production of electronic components fell -13.8% YoY in April, the 6th consecutive decline with the downward trend accelerating. As this is just domestic production, this data alone is inconclusive but, on the margin, it’s getting a little worse, not better. But where we are seeing clear weakness that is worse than 6 months ago is machinery.

​

We met with the largest wholesaler of machinery equipment in Japan. They see orders falling across the board. They acknowledge that automotive companies appear to be doing better, but that is more due to destocking of inventories that they had slowly built up at the OEM level as the supply chain had begun to loosen. It has not been followed up by any machine order pickup. While some smaller auto parts manufacturers had a slight uptick in orders in May, it is still at very low absolute levels and, outside of autos, general machinery orders are softening, not strengthening. As the domestic construction market continues to be soft, construction machinery is also weak although exports to North America continues to be firm. One bright spot is the US commercial construction market where, despite regional bank risks, orders remain brisk. Similarly, US infrastructure construction demand appears to be strong too. But otherwise, whether it be other countries or other industries, we hear very little positive datapoints from an end-industry perspective. Machine parts orders (i.e. those that sell into these machinery companies) are also seeing declining orders. National statistics are saying the same thing with machine tool orders falling for 5 straight months in a row (9 months for domestic orders). As if that’s not enough, monthly consumption of cardboard boxes by the machinery and electronics sector has fallen for 8 months straight and, in the last 2 months, has been below even 2020 levels.

 

Now, one might think that the machinery cycle must, then, be close to the bottom. But if one looks at the absolute levels, orders are around mid-cycle in Japan and the US (local currency terms) while still above mid-cycle in China and Europe. The only region below mid-cycle is Asia ex-China (notably Thailand). If history is any guide, while the gamma may flatline soon, the delta will stay negative for at least another 10 months (given the shallower speed of the down cycle than in the past, I’d bet it might take a little longer). I’d note that this is orders, not revenue (nor this year’s profits). Lead times are much longer than in the past and backlogs are only just starting to be worked down. While labor and material costs are higher, most companies started price hikes from mid last year which will, from an YoY perspective, look better in the front half. And the weaker yen will most definitely be supportive for industrials whose majority of sales is still exports. As such, sales and profits will be fine, particularly this quarter. They may even look strong, especially with the yen at above 140 (most companies probably guided using a 125~130 yen forecast, given that’s where it was in Feb~Mar when forward estimates were being made). Perhaps the current decline in orders will start to pick up before the backlogs are worked down such that sales and profits will stay firm in the second half and throughout next year. With the longer lead times, it’s certainly possible. But the cycle would have to thread the needle perfectly (in a period of rising global interest rates). In my long experience, orders typically lead. Maybe this time is different. The market certainly seems to think so. But I think I’ll stick with what I know rather than chasing ghosts. When and if the ghosts materialize, I’ll accept that this time was different.

​

Masaki Gotoh

​

===

​​​

“Mathematics is not about numbers, equations, computations, or algorithms: it is about understanding.” – William Paul Thurston, American mathematician.

​

I probably mentioned it once before, but a few years back, I took my children out of the Japanese public education system and transferred them to an international school. I originally wanted my children to go to a normal Japanese public school but having never been educated in Japan myself, I wasn’t prepared for what, or rather, how they teach. It took me quite by surprise and I immediately became disillusioned by it. I love this country intensely … but when it comes to education, it is (at least in my Western view) disappointing. I’m very happy with my choice and they are getting the elementary school education I hoped they’d receive, although there are a few drawbacks. We still live here in Japan, so they still have to learn basic Japanese and, sadly, they are at least 1~2 years behind their public-school peers. They don’t have to become the next Japanese Shakespeare or anything, but I’d like them to be able to get around and about without any language difficulties. The other is math. I don’t know if it’s the school we chose or the international schools in Japan in general, but their mathematics education is embarrassingly slow. And so, on weekends, in addition to their weekend homework, I have them work on, among other things, Japanese Kanji drills and at least Japanese grade level math, if not higher.

​

As the textbooks are Japanese, the text teaches as they do in Japanese schools (which was what I wanted to avoid in the first place). Therefore, it becomes rote learning-based where equations and computations are crammed into complicated, meaningless questions that helps one compute (quickly), but not necessarily understand. Worse yet, these are supplemental texts for children already attending Japanese schools, so they aren’t meant to teach but to reinforce. And so, I spend most Saturday and Sunday mornings teaching them the “why” of math. Yes, you do need to learn and remember the equation, and I’ll even teach them a few tricks to compute faster (they’ll have to take some sort of entrance exam someday so learning to solve faster can’t be a bad thing). But they have to understand why this equation solves the question. For example, a few weeks ago I explained to my son why, given “the ratio pi = circumference of a circle / diameter of a circle”, “pi times r-squared” solves for the area of the circle (the text, of course, doesn’t explain it). Once you understand the why, everything else clicks. Anyone who’s taken basic options class knows that they don’t just tell you to learn the Black-Scholes equation and be done with it. You’d be finished after one class if that were the case. One must understand random walk and Brownian motion before one can derive and understand option pricing theory which gets you to the basics. But with that knowledge, one can expand their knowledge to price a variety of different derivatives and could help one get a job at Goldman Sachs (I was head of equity derivatives research for Asia-Pacific back in my Goldman days for a brief period of time).

​

Understanding the “why” of something as logical as mathematics also has the advantage of sometimes asking very simple, basic questions regarding new topics. For example, as I often say, I am no macroeconomist. It was something I never really had to seriously think about in my investing career until the last 3 years or so. I’ve taken basic Economics classes in business school (I hadn’t in college as I was a computer science major), so the terms don’t frighten me although I thought, no offense to the economists out there, that the topic was rather boring. I enjoyed Strategy and Marketing much more (ironically, the class I hated the most was Accounting, but that’s because, like Japanese education, it felt like rote learning without asking the “why”; but in this career, it’s probably the most important topic after Finance). But back to basic macroeconomics, I never really had to think it through. So perhaps someone can help me with some basic understanding:

​

Why is the world obsessed with YoY inflation figures? Sure, I can understand why, depending on who is asking, one might want to care about the headline CPI vs the core CPI or, nowadays, core services. But is it just the YoY that matters because that’s all that people seem to discuss? Looking at the basic US CPI Urban Consumers YoY (non-seasonally adjusted) headline number, the line is rapidly falling (all charts below are from Bloomberg).

​

​

​

​

​

​

​​

​

​

 

 

 

 

 

So, inflation is rising a lot slower than it was a little while ago. It does look like, in a few months’ time, that this line will be back near where it used to be. That’s good. But is the consumer happy? I’m guessing that the consumer is looking at the broader CPI since they have to eat and drive and buy goods in addition to paying for services, so I’ll stick with the broader CPI. Here is the same CPI nominal figure across the same time horizon.

​​

​

​

​

​

​

​

​

​

 

 

 

 

 

 

That acceleration since end 2020 is what concerns me. This indexed figure was around 260 as of the end of 2020. It’s now at 304. In 2.5 years, the cost of everything has, on average, risen +16.7% (in the US). Assuming the historical trendline before 2020 is the rate that US inflation should be rising (whose slope looks fairly linear), even if I were generous with the end points to estimate the slope, it would take about 4 years to get back to the original trendline if MoM inflation were 0.0% every month for the next 4 years. Or maybe wages are also rising faster?​

​

​

​

​

​​

​

​

​

​

​

​

​

​

​

The source is the same as the CPI figures (Bureau of Labor Statistics) showing US average hourly earnings private sector (seasonally adjusted). Yes, it does seem like it’s accelerated a little since end 2020 compared to the previous decade, which coincides with the tight US labor market. This indexed figure is up about 11.7% since end 2020 (compared to +10.7% for the previous 25 months). So it’s up, but clearly slower than inflation while wages before 2020 were generally rising faster than inflation.

​

Now I have no idea how these figures are computed or how accurate they are. I may be looking at the wrong statistics. But I get the feeling that the consumer isn’t as happy as the stock market (except, of course, those that are invested in the stock market). They are either earnings less on a relative basis or working longer hours to keep up which doesn’t sound like a particularly fun place to be. Additionally, the interest on new loans will be at least 3% higher than it used to be (or more if variable rates). According to the Fed, total household debt reached $17.05 trillion as of 2023Q1 vs $14.56 trillion as of 2020Q4. Per capita, it’s risen from $52,681 to $60,249 or +14%. The last time it rose more than +5% per annum was pre-financial crisis. And, we also have to remember that student loan payments will restart this fall.

​

Now, let me say again, I am no macroeconomist. And because of that, I’m certain I must be missing something because, when it comes to this topic, I’m certain to be an absolute novice. The US consumer has typically led the world economy and, from the 2010s, the Chinese consumer became an additional driver. While none of this really effects my day-to-day or how we invest in companies, I try to be aware of how the US and Chinese consumer are feeling and, for my domestic names, the Japanese consumer as well. Right now, it appears that the Chinese consumer is taking a cautious approach to normalization while, according to recent statistics, the US consumer is still very resilient. So even though we are all, on the margin, poorer than we were 3 years ago, maybe the tight labor markets will keep Americans spending.

​

But I’m just thinking back to basic math. If something, say stock prices, falls 50%, it would take a 100% return to get back to where it was. When we say we are looking for a 30% margin-of-safety, that’s a +42.8% return if it gets back to fair (1 divided by 0.7 = 1.42857). That’s just math. Similarly, if something like, say, inflation, were growing at about 3% per annum (or more like 2% in the last 10 years before the pandemic, but that was probably abnormally low) and it temporarily rose to a figure well above 4% for 2 years straight, that figure has to fall to 0% (or less) for a fairly significant period of time to get it back to where it was. Simply getting back to 3% doesn’t really help the consumer feel as good as they did before unless wages accelerate, I would think? So, in order to do so, I’d think the Fed would have to keep rates fairly high for a sustainable period of time even if inflation rates were to fall to get back to equilibrium (just like they’ve been continually saying but the market appears to have ignored until recently)? I’d be grateful if someone who understands this stuff much better than I explain to me where my logic is flawed.

​

​