Dealing with Turbulence

Summer 2023 was a reminder that stock markets don’t just go up. Our portfolio dropped, though less than half a percent, while market indices as a whole dropped more.

Underneath the surface, we saw some company-specific bad news in our portfolio. Some stocks fell due to gravity, it seems. Axcelis, Arlo, and Apple all dropped after strong starts to 2023. But some was company-specific problems*. Those are the most interesting and challenging things to deal with as an investor.

*Unlike last quarter, where we saw big but random sell-offs that quickly reversed.

Two of our biggest positions are in Progressive Corporation (PGR) and Discover Financial Services (DFS). They are S&P 500 components and well-known financial companies. Progressive is the largest auto insurer in the U.S. sales wise, and considered a leader. Discover is a credit card company and bank. It is usually lumped in a second tier with Capital One and Synchrony Financial, behind American Express. (Visa and Mastercard are payment networks, not credit card companies per se).

The Bad News

Progressive and Discover had bad news in July, of differing import and impact. Progressive’s June earnings report, released in mid-July, was bad. The company reported a combined ratio of 104.9, vs. 94.7 a year prior. The combined ratio is a measurement of how much money an insurance company makes on its actual insurance policies. If it’s over 100, the company is losing money. Progressive attributed a lot of this to a reserves adjustment on past claims. PGR shares dropped over 13% that day.

Discover Financial released its Q2 earnings a week later. The earnings themselves were fine. Discover missed estimates marginally, but for financial companies, earnings and revenue estimates are less precise, I’ve found. The real negative news was the discovery of a misclassification of certain credit card accounts dating back to mid-2007. This was framed as relatively small – the company prepared to refund up to $365M, vs. income of $901M in the second quarter. It still had an outsized effect for two reasons. This is the second consecutive year Discover has reported a compliance or accounting issue in its summer earnings report. And as with last year’s report, Discover chose to suspend their share buyback program, which is one of the major drivers of earnings and value for shareholders. Discover shares dropped nearly 16% that day.

Neither of these issues were wholly predictable, nor wholly surprising. Progressive and the auto insurance industry has been struggling with inflation of used car costs. Repairing or replacing cars after accidents has gone up considerably. And Discover had the same sort of issue a year ago, related to its student loan business. The line goes that there’s never just one cockroach. This isn’t quite a fraud-level ‘cockroach’, but it’s not great. At the same time, you can’t really foresee these things with precision, which leaves us to react.

How we reacted

We didn’t sell any shares of either company, and in fact bought shares in both, but at different times. In Progressive’s case, we increased our number of shares by about 15% in the immediate sell-off. The earnings problem was less surprising. The reserve adjustment is just a different flavor of ‘it’s expensive to resolve these claims’ that has been weighing on Progressive since 2022. Florida has been a problem. Progressive is likely to either raise prices enough to cope with it, or see auto repair and the Florida issues calm down, or both. Progressive has been taking rates – its term for raising prices – and is not alone in facing these challenges. This reassures me that this is something the industry will overcome. The subsequent two months of earnings* produced combined ratios of 90.6% and 97.2%, vs. Progressive’s goal of 96. Quick progress, if perhaps just coincidental.

*Progressive is unusual in that it reports earnings every month, not just every quarter.

In Discover’s case, we increased our number of shares by about 10% in the last week of September. This came just above Discover’s new 52-week low, and another 17% below its post-earnings price. I waited not so much because of the issue itself, but because it’s a specific to Discover problem and we don’t quite know where the end to it is.

Or at least, we didn’t in July; in August, the company announced its longtime CEO would be leaving. While management called this a planned exit, there is no permanent CEO taking his place yet. At the same time, the interim CEO led a conference call where he stated that:

1. There were no other major shoes to drop, and;
2. That the share buyback should be returning sometime soon.

This makes me think the problem is more an ‘added cost of business’ than existential risk. Discover will spend more on compliance. This will hurt its profits in the short term. But it won’t draw crippling fines or restrictions. News on the last day of the quarter underlined this. Discover signed an FDIC consent order requiring them to up their compliance game, but not to pay any fines. The stock rose 4.8% on the following market day, a sign of how much worry there is in the stock.

A good hire as CEO and a restarted share buyback should both be boosts to the stock. At 6x earnings, we have at least some margin of safety for anything else to go wrong (like increased delinquencies as customers struggle to pay back credit cards, even beyond what is already happening).

Bad news happens, each time in its own way

There are bad news events that are enough to push me to sell all of my stock. But usually, it’s more of a ‘straw that broke the camel’s back’ story. We sold our last shares of Wolverine Worldwide (WWW) after the company suddenly fired its CEO, who had only been there for three years. The company’s decision is probably a good one. WWW got rid of an outsider and promoted the COO, who has been in the company for decades. But it’s also a sign of how tough a road the company has in front of it, and the challenge of being a 2^nd or 3^rd tier player in footwear.

In PGR and DFS’s cases, we’ll see. I think they both have competitive positions, brands, and valuations that should work out favorably to shareholders. The bad news reminds me that not everything will go right. But in the market, we end up making the most money when we can distinguish between which bad news offers an opportunity, which triggers a ‘wait and see’ mode, and which is a fire alarm.

I should add, posting this on October 6^th, that we’ve had another bad news event in the portfolio. As seen below, two Mexican airport companies are meaningful positions for us (8.4% of our portfolio as of September 30^th). They announced that the Mexican government unilaterally changed the terms for how much they can charge customers. This caused a sell-off of ~24-26% on October 5^th, with it being much worse at times during the day.

This is a bad news event because:

1. The companies will be less profitable in almost all scenarios (the exception: if they sue in court and win, but even then their contracts for maximum tariffs will be renegotiated in the next couple years)
2. It is a sign that the Mexican government is becoming less predictable for investors, with an election year coming up
3. We don’t yet know what the impact is (all three publicly traded Mexican airport companies said “we’re still studying the impact”).

The first reason is bad but calculable if we figure out the other reasons. I don’t know more yet. We’re filing this as a “wait and see” with a chance of being either an opportunity or a fire alarm, depending on how big an impact the changes are, and what it says about working with the Mexican government.

For context, that day wiped out most but not all of our relative gains vs. the S&P 500 and Nasdaq in Q3. We’ll see where we go from here.

President Andreas Manuel Lopez Obrador, reminding us to stay humble. Source: El País

The underlying point: there are always risks in stocks we own. If there weren’t, the risk would be that the stocks are too expensive, because everyone would buy them. Government risk is a typical top risk with infrastructure companies that depend on governments, especially outside the U.S. And Andreas Manuel Lopez Obrador, the current president, has been a populist left-wing (ostensibly) president, and thus anti-business. He will be out of power next year. Either his successor or a more business friendly candidate will be in charge. We will see how they will handle these sorts of relationships, and whether the last nine months or so of AMLO’s term offer further surprises.

Market Outlook

All major U.S. (and European) indices dropped, as did our portfolios. I could point to the Federal Reserve continuing to insist it will not cut interest rates in the near-term future. This has created a Lucy and Charlie Brown dynamic, except Lucy (the Fed) has repeatedly told Charlie Brown (the market) that she will pull away the football, and Charlie keeps running at it anyway. The excitement over AI has dulled a little too. It’s become clear that the spending on AI won’t immediately translate to profits for suppliers, nor better products and services for spenders.

But ultimately, the S&P 500 rose 24% in the prior 9 months. It’s not very notable that it dropped less than 4% this quarter, with the drop all coming in the last two weeks of September.

Our Q3 performance

	Q3 2023	YTD 2023	2022	2021	2020
U.S. portfolios	-0.3%	33.5%	-13.4%	16.8%	12.0%
S&P 600	-5.4%	-0.5%	-17.4%	25.3%	9.6%
Russell 2000	-5.5%	1.4%	-21.6%	13.7%	18.4%
S&P 500	-3.6%	11.7%	-19.4%	26.9%	16.3%
Nasdaq	-4.1%	26.3%	-33.1%	21.4%	43.6%
European Portfolios	-3.6%	4.1%	-15.3%	4.5%	18.3%
Euro Stoxx 50	-5.1%	10%	-11.7%	21.0%	3.5%
DAX	-4.7%	10.5%	-12.3%	15.8%	-6.3%

See Disclaimer at end of post.

Key Portfolio Stats

• Our portfolio level price to earnings for trailing 12 months (TTM) was 19.35*, or a 5.2% yield. Our price to free cash flow TTM ratio was 14.63, or a 6.8% yield. This compares to 21.19 and 17.38 P/E and P/FCF at end of Q2
• Cash and equivalents (the ETFs MINT, JPST, SGOV, and BIL, and 13-week U.S. treasuries) was 19.8% of our quarter end portfolio, with an average yield of 4.6%. This compares to 21.6% of our portfolio and 3.9% yield at the end of Q2.
• We bought 35% more equity positions than we sold in Q3. For 2023, we sold 3% more equity positions than we bought.
• We’re at a lower cash position than I have had for some time, for three main reasons: some withdrawals at accounts for personal reasons; I think we should be a little closer to fully invested as a normal baseline; and our VMW position will turn into cash and AVGO shares at the end of October in all likelihood (see below).

*I adjusted earnings for impairments on AER in the last 12 months; took FG’s adjusted net earnings; and free cash flow is a shakier number due to the number of financial companies we own. I have not factored in that VMW’s shares are basically a fractional right to AVGO’s earnings. The numerator is the total value of our long equity positions as of the end of Q3.

Our ten biggest equity positions at the end of Q2 were:

• Axcelis (ACLS) – 9.9% of our portfolio
• Progressive – 6.3%
• Grupo Aeroportuario del Centro Norte (OMAB) – 5.5%
• Dropbox (DBX) – 5.5%
• F&G Annuities & Life (FG) – 5.1%
• VMWare (VMW) – 4.6%
• Discover Financial – 4.3%
• Atkore (ATKR) – 4.1%
• Arlo Technologies (ARLO) – 4.1%
• Booking Holdings (BKNG) – 3.5%

Top winners (performance as % of starting portfolio level for quarter):

Steelcase (SCS)	1.1%
VMW	0.7%
FG	0.6%
BKNG	0.5%
PGR	0.4%

Top losers (performance as % of starting portfolio level for quarter):

ACLS	-1.2%
DFS	-0.9%
Apple Inc (AAPL)	-0.4%
Juniper Networks (JNPR)	-0.3%
Taiwan Semiconductor (TSM)	-0.3%

Notes on Winners/Losers

Steelcase (SCS) was our big Q3 winner. The office furniture company had a solid earnings report, beating estimates for the quarter while also raising earnings guidance for the year. It is reducing inventory, recovering gross margin, and still well positioned for if/when an increase in office furniture demand comes.

The weak point in the report was orders, which were down year over year. The company reduced its revenue estimates for the year as well. Efficiency is great, but if the industry isn’t growing, it probably won’t help. For now, though, it seems like Steelcase is on the right path.

I rectified some of the mistake in selling VMWare shares prematurely last quarter, by buying them back in a couple accounts. Broadcom (AVGO) is buying VMW, and a VMW share basically consists of $71.25 + .126 Broadcom shares*. That means VMW shares should be worth $175.9. They closed trading on September 29^th at 166.48, offering us a 5.7% arbitrage spread. All things equal, we will make 5.7% for just holding VMW shares, though Broadcom’s value is a moving target.

* Technically, VMW shareholders can choose to receive $142.5/share or .252 Broadcom shares / share, and then Broadcom will pay for the full deal with 50% cash and 50% Broadcom shares, pro-rated to all shareholders based on their elections. In practice, .252 Broadcom shares are worth way more than $142.5, and postpone tax payments, so I can’t imagine anybody who will opt for cash, meaning we will all get half and half.

That spread was 25.6% on June 30^th, and narrowed because the merger deal got approvals or no complaints everywhere except, so far, China, and all the reporting out of China is good. The deal is due to close October 30^th, in time for the end of Broadcom’s fiscal year accounting-wise.

Broadcom is one of the biggest semiconductor companies in the world. It bought several semiconductor companies in the 2010s, and then shifted to buying older generation software companies – CA Inc, Symantec’s security business, and now VMWare, which is a geriatric millennial of a company. Broadcom’s business has expanded from providing parts to iPhones to supplying a whole range of modern semiconductor uses, including AI chip development.

Its shares are fully valued – a PE of 25.5 before accounting for VMWare, a price to free cash flow of 17.6x if you incorporate VMWare. The company and CEO have a good track record and reputation, and the company’s balance sheet is solid despite all its acquisitions. It pays a big dividend.

I sold the three shorter-term positions in VMW after making money on the closing spread. We’ll hold onto our Broadcom shares once the deal closes – and I would be stunned if it does not – through at least year end, to postpone any taxes from selling them. After that, we’ll see how it’s priced and how it fits into our already semiconductor and tech-heavy portfolio.

New Positions

Machten

We opened a small (.6% of our portfolio) position in Machten, a newly independent telecommunications company based in upper Michigan. The position began via a spin-off from LICT, a bigger but also small telecommunications company, which wanted to divest Machten to focus on its “west of the Mississippi” businesses, and to be done with the “undue time” spent on Machten. We then bought more towards the end of the quarter.

Machten sells broadband, phone, and video access in much of the upper peninsula, a few sporadic counties in the lower peninsula, and the Traverse City area. Telecommunications companies don’t grow very much – landlines especially erode over time. Machten’s revenue year over year is down. It doesn’t, on the surface, seem very attractive.

I bought shares for a few reasons, though. First, they’re cheap – Machten shares, using the average price at which I bought them, trades at just over 5.1x EBITDA, an earnings measure that is useful for telecom companies, and 8.3x last year’s earnings. In my experience, good things happen to telecom companies at these prices. There haven’t been many acquisitions since Alaska Communications got taken out in 2021, but if Machten can hold its business steady, it should be fine.

It should be able to hold its business steady because a majority of its revenue is from government programs to support rural broadband deployment. Supporting rural communities is a bipartisan issue. While the funding is only earmarked until 2028, it is likely to continue beyond that. Machten is also growing its non-regulated revenue, a sign that it’s able to compete in offering broadband service.

The last, more speculative reason to own Machten shares: I believe Northern Michigan will become more attractive as a place to live over the next decade plus. As climate change continues to have an impact, Michigan’s colder climate will become more attractive both absolutely and relatively. Its access to the Great Lakes water system will be a strategic plus. This doesn’t mean that Machten will benefit – there are major telecom players in the area. But, it’s a nice backdrop for owning a cheap, sleepy company.

(I would have visited their headquarters on a trip to Traverse City this summer, but I misremembered the address and was a block away and with a dead cell phone battery).

Levi Strauss

We opened an even smaller starter position (.45% of our portfolio) in Levi Strauss.

Levi Strauss is of course the global jeans maker, though it does sell other clothing – tops, khakis via Dockers, and athletic wear via recent acquisition Beyond Yoga.

Despite a 150-year history, the company has only been publicly traded since 2019, along with a stint in the 1970s and 80s. Levi Strauss’s descendants still control the company’s voting rights, meaning they have ultimate say over the company’s future (a sale of the company, for example, is less likely).

The company has a solid balance sheet, with net debt of 1x their EBITDA (earnings before interest, taxes, depreciation, and amortization, meaning their raw cash generation before they reinvest in their business) and really low interest rates on their debt through 2027. Its profitability is ok with room to improve. And the shares are priced cheaply – 11.5x analyst estimates of 2023 earnings, and 9.6x 2024 estimates, at the price we bought shares.

Why is it cheap? First of all, most apparel companies I looked at that are not lululemon or Nike are cheap. It’s not only Levi Strauss. Many of these companies carry high inventories, which is a risk because as fashions change companies are forced to mark down inventories at low prices, meaning they sell at a low profit or even a loss.

There’s also the market fear that consumer spending is finally slowing down. The return of student loan payments and general angst about the economy, plus spending power that can’t go on forever (can it?) are deemed the likely culprits.

And longer-term, I believe there are doubts about brand power in general. How much do old-world brands matter when a Mr. Beast or a Kardashian or a TikTok star can build up a huge audience and then monetize it in a number of ways, including clothing? Levi Strauss will have a new CEO next year, Michelle Gass. Unlike Discover, the transition process seems very deliberate and thought out. She will have been at the company for over a year before she takes over as CEO. But her last role, as CEO of Kohl’s, wasn’t a great success. Even if you don’t blame her for the hand she was dealt, the question is: is investing in a retailer in the 2020s like investing in a department store in the 2010s?

My track record investing in retail companies – bad – and my feeling I don’t totally understand are what’s keeping this a tiny position for now. Levi’s earnings came out October 4^th, and were not great. Inventories went higher (though I suppose in advance of the holiday shopping period), earnings missed, the company took an impairment on Beyond Yoga. The management team sounded optimistic on the call and tells a nice story, but there’s still a lot to prove. I’m not excited to add any at this point, and may decide to close our small position for now and resume watching.

Closed Positions and other notes

• The only position we closed was Wolverine Worldwide, as discussed above.
• We reduced our positions in Axcelis, VMWare (after buying back in), Apple, Arlo, Dropbox, Booking, Grupo Aeroportuario del Centro Norte, the RWM ETF, and Juniper.
• We added, besides the names mentioned above (PGR, DFS, MACT, and LEVI), to Aercap, Atkore, Dallas News, Mosaic, and Charles Schwab.
• We traded out and then back into Bassett Furniture, and I held a position in U.S. Steel for a day in my personal account, losing a little bit of money on a merger bidding war bet.
• On Aercap, I’ve gone back and forth with this stock over the past year. On the one hand, aircraft lessors are more in demand because airlines are not willing or able to buy planes for their balance sheet as much as they were pre-pandemic, and because aircraft production is still not at full capacity. And I believe in the decade-long travel tailwind. On the other hand, Aercap’s stock has faced two crises in three years – the pandemic and then the Russia-Ukraine war. And because it’s the biggest aircraft lessor in the world, it’s hard to imagine a company buying it out, unless a Berkshire Hathaway type player gets interested.
  
  I’ve come down on the side that Aercap is always going to trade at a discount to its book value – how much its planes are worth minus all the company’s debt, essentially – and that it should be able to survive crises as it has this time around.
  
  After I came to that decision and added a few shares, AerCap announced it was getting money in an insurance settlement related to its Russian planes. This is a good sign that it might win more of its insurance claims, proceeds of which it would then use to buy back its own shares, as it did this time. As an added bonus, GE sold Aercap shares it owns, reducing its ownership to around 15% of the company. GE’s need to sell is another artificial cap on AER’s shares, and should be gone in the next year.
  
  The last point here is that AER is pretty consistent at growing its book value by 10-15% every year. We now have three things to own shares for
  
  1. That steady growth
  2. The wider than necessary discount to book value (it closed the quarter at .84x my estimate of AerCap’s book value, vs. the .9x I think it should trade)
  3. The potential insurance claims, which would amount to another $12.9/share if it comes through.

Disclosure: I am long or short all positions as mentioned in this letter, plus Berkshire Hathaway B shares, and minus Bassett Furniture, which I sold. I may change positions at any time. I have no immediate plans to make major changes. This is not investment advice. Investing is risky. Any investing decisions are your own responsibility and should be taken after speaking with an advisor or at your own risk. This is not a solicitation to buy or sell anything. Past performance is of course no promise of future results.

Disclaimer: I calculate performance and all portfolio figures myself, manually, so it may be prone to error. The accounts I manage may deposit or withdraw money over the course of a quarter. I account for that in my calculations by adding/subtracting that money to/from the starting amount at the beginning of the period. This means withdrawals intensify performance and deposits dampen it. For half-year, 9-month, and full-year performance, I multiply quarterly performance by one another to control for deposits/withdrawals. If there’s a better way to calculate, please tell me!