What I Hate and Love about Singular Diligence

Recently I went out to a restaurant I like. Unfortunately, the restaurant went out of business. That reminded me how tough it is for small business to survive and led me to some thoughts on Singular Diligence.

It’s been 1.5 years since we launched The Avid Hog (the predecessor to Singular Diligence). Geoff and I actually worked on the newsletter for 3 years. Somehow we still survive, building up an archive of well over 1,000 pages of research and notes. Only passion and perseverance can lead us this far. And there are things I hate and love about the newsletter.

I simply hate the pressure. Geoff and I want to work together. He had the idea of writing a newsletter in early 2012. I never thought it’s possible. Warren Buffett says he’ll “settle for one good idea a year”. How can we come up with a good idea a month?

Make Priorities

We must make priorities. We give priority to downside protection over upside. So, clients won’t lose money if they act today. We also give priority to business quality over cheapness. Business quality doesn’t change fast so a qualitative research can be timeless. If clients read our research today, they’ll be ready to act quickly in the future. Clients will make the best return if they see the newsletter as a tool instead of as investment advice.

Choose Candidates Better

In the early days, our focus was on improving the research process. We frequently ran into “crisis” when we don’t have the next stock to analyze. There were also times when I realized that a company isn’t good enough only after 2 or 3 weeks of research. I had to drop the stock. So, it’s a big risk to choose a wrong candidate.

We started building a process to maintain a candidate pipeline in September 2014. Geoff and I have a candidate meeting every week. We look at our watch list, screen, or other blogs to find ideas. We try to pick the best 2 or 3 stocks to discuss about in the meetings. We focus the discussion on risks and normal earnings. It’s important to have financial data for these early discussions. So, I type data from all 10-Ks of each stock into an excel template. This is a powerful process because we discuss about 100-150 stocks each year.

We maintain a list of top 10 candidates. The unbreakable rule is that a stock must be below 15 times EV/After-tax Normal Unlevered Earnings to appear in the list. The rule makes sure candidates trade at a below average price. This means we won’t pick stocks where we love business quality and prospect so much that we compromise on price. The rule makes our job difficult in today’s environment when people are talking about new low normal interest rates. Sometimes we have only 8-9 candidates in the top 10 list. But it helps push ourselves to actively search for new candidates.

To reduce the risk of choosing the wrong stock, we do a lot of basic research on top candidates. We do scuttlebutt in this stage. Some clients help us do scuttlebutt by talking to employees or learning about customers and products. Afterwards, we can contact the management if necessary. We also read Investor Day transcripts. My experience is that analysts tend to ask short-term oriented questions in quarterly earnings call. There’s not much useful information there. But there’s usually great information in Investor/Analyst Day conferences. So, instead of reading 1,000 pages of earnings call transcript, I read only 50 pages of Investor Day transcript in this stage.

Price Movement Is a Headache

Our weakness is long lead time. There are months from the time a stock enter the top 5 candidates to the time I start analyzing the stock. I do research for a month and send my notes to Geoff. Geoff does further research and writes the final report in another month. A stock can stay in our inventory for several months because we have some restrictions. For example, we try not to write about two obscure foreign stocks in two consecutive months because that upsets clients who prefer buying U.S. stocks. We can’t write about two micro-cap stocks in two consecutive months because clients may hate illiquidity.

We’re subject to price movement because of the long lead time. It’s heart-breaking to see the price moves so much that we’re unable to publish a report. Some examples are PetSmart or Greggs. Sometimes the stock price increases by 20% even before I finish my notes. I just talk to myself “no, I hate this job.”

One solution is to increase the speed we do research. I think we’ve increased our productivity by triple-digits since when we started. But we just reinvest productivity gains in further depth of research. Three years ago, it took me a month just to read and analyze all the information I can find about a stock. Today, I learn also about competitors, customers and suppliers.

Another reason that restrains our pace is procrastination. It’s always tempting to read some more instead of starting to write. So, I don’t think that we can analyze a stock in less than a month. And price movement remains our biggest risk.

That’s what I hate about Singular Diligence. People say that if you choose a job you love, you’ll never have to work a day in your life. But when you have to race against the deadline and struggle to make money, it’s not fun at all. Sometimes I just hate what I like to do.

The Best Way to Learn Is to Practice

But there are good reasons for loving this job. First, I think this is the best way to learn. I have a college friend who loves investing but doesn’t read investment books. He said that reading investment books doesn’t teach us how to make money because otherwise the authors wouldn’t share. I agree that we don’t become a better investor by reading books. But books prepare us to become a good investor. We get knowledge from books. And we start learning when we practice.

I always learn something new from each research. I can practice techniques that I learned from books. One example is we all learn that there’s a red flag when inventories grow faster than sales. I see that at Swatch (VTX:UHR). In 2014, sales grew 3% while inventories grew 10%. In 2013, sales grew 8% while inventories grew 23%. Is it a short sign? The answer isn’t that simple.

Tom Russo talked about working capital in a recent interview. He said that Wall Street analysts and some activist investors usually prefer lower working capital and higher cash flow. But long-term investors welcome more capital being reinvested in a great business.

I find that Swatch’s inventory turnover declined consistently from 3.1 in 1998 to 1.5 in 2014. There can be some fundamental changes. Swatch might have grown retail operations in this period. They might be investing in infrastructure in emerging markets. So, it’s useful to talk to management about this topic. It’s necessary to compare Swatch’s inventory turnover with Richemont’s during this period. The two companies are different so we should also compare finished inventory turnover.

So, the actual exercise is much more complicated than the technique we learn in a book.

Things that Books Can’t Teach

There are also things that books don’t teach us. Books about moats can’t help us analyze Majestic Wine (MJW: LN). They don’t have the purchasing power of supermarkets like Tesco. They can be killed by online competitors. But analyzing Majestic Wine’s business model and cost structure can tell a different story.

Similarly, I never read a good book that teaches us how to estimate normal earnings. That’s because each actual situation requires a specific approach.

When Babcock & Wilcox (NYSE:BWC)’s government contracts are in serial production, they do the same thing every year. Costs are visible so past margin can be a good benchmark for future expectation. But we must be careful if they do a prototype project.

Swatch is run by a long-term oriented management. They’re willing to spend today to grow their portfolio of watch brands. Moreover, EBIT margin is cyclical but has a long-term upward trend. What should we do? Should we use current earnings or peak earnings? Should we estimate earnings using EBIT margin? If yes, how do we estimate the normal margin? A good answer requires us to understand why margin increased and have some expectation about future revenue.

Sometimes we have to delve into product economics to estimate normal earnings. America’s Car-Mart (NASDAG:CRMT) sells and finances used cars in small towns in South-Central states like Arkansas, Oklahoma, and Missouri. They don’t lend money but cars. The investment laid out isn’t in receivables on the balance sheet, but in the cars that they lend to customers. So, sales and EBIT margin are meaningless.

Encore Wire (NASDAG:WIRE) is another example. Encore Wire makes copper wire. The copper cost is almost 70% of revenue. So, the copper wire price mostly tracks the commodity copper price. Copper went from less than $1 per pound in 2003 to over $3 per pound today. So, it’s hard to estimate Encore Wire’s normal earnings by looking at sales and margin. The better approach is based on EBIT per copper pound of product.

Familiarity with One Company Can Help Analyze Another Company

I also love the chance to study a new industry each month. I can learn about nuclear components for one month and then move on to the most comfortable recliner in the world the next month. I can also develop new interests in wine or mechanical watches. That does help reduce the stress of racing against the deadline.

More importantly, familiarity with a company/industry can help analyze another. Studying the decline of the Omega watch brand in 1970s and 1980s can tell us to pay attention to the management when we analyze Ekornes (EKO:NO).

Familiarity with Coach or some other luxury brands can help us understand the distribution side of Swatch’s business. It also teaches us about the importance of scarcity in luxury.

Sometimes the connection is indirect. Reading about George Risk (RSKIA) shows the tendency of customers to re-order from the same supplier if they deliver on time. That can gives us some clue to the behaviors of contractors who buy copper wire from Encore Wire.

I see some similarity between Progressive (NYSE:PGR) and Encore Wire. About ½ of Progressive’s business is sold through the direct selling channel like Geico. Direct sellers have durable cost advantage because competitors are stuck with the agency channel.

Encore Wire is the second largest manufacture of electric wire in the US. They started in 1989. Today they have about 25% market share. They grew organically. They never made an acquisition. All earnings were reinvested into the manufacturing complex in McKinney, Texas. They have only one distribution center there. Their order fill rate is over 99.9%. They deliver to all customers within 7 days. They offer customization like selling 825 feet of copper wire instead of the 1,000-foot standard put-ups.

Encore Wire’s competitors tend to grow through acquisitions. They can have 3 or 5 plants and 10 distribution centers (DC). They ship from multiple plants to multiple DCs, and from multiple DCs to customers. So they have to handle shipping many times. Competitors’ shipping cost as % of sales can be 3% to 6% higher than Encore Wire.

Encore Wire also offers better service. Competitors are closer to customers. But there are 10,000 SKUs. They can’t hold full inventories in all 10 distribution centers. For an order, they can fill 50% from the nearest DC, 30% from a farther DC, and 20% from an even farther DC. They can take weeks to fill order. And it’s difficult for them to offer customized orders.

Encore Wire has the lowest cost in the industry but has the highest price because they offer great service. Meanwhile, competitors are stuck with their distribution model. They can’t close plants and concentrate on just one like Encore Wire.

Both Progressive and Encore Wire are the low cost players in their respective industries. They are both very cautious in pricing. They both do best when there’s cost inflation.

Another example is that understanding QLogic (NASDAG: QLGC) can help analyze Breeze-Eastern (NYSE:BZC). QLogic makes fiber channel adapter. Fiber channel adapter is a critical component in a storage-area network but is a tiny portion of the total cost. It takes a long time to go through OEM qualification and testing. QLogic has about 54% market share.

I see the same thing at Breeze-Eastern. They make rescue hoist. This is a mission critical component in a helicopter that pulls people up and down in search and rescue missions. The qualification process with a new airplane model is incredibly expensive and time-consuming. Breeze-Eastern has about 50-60% market share.

The Fountainhead

The last reason I love this job is simply I love value investing and I like to work with Geoff. I draw inspiration from Howard Roark in my favorite novel, The Fountainhead. Howard Roark is a visionary architect. He has his own philosophy. He works with his mentor Henry Cameron because they share the same philosophy. Their designs are unpopular and they make minimal amount of money. But they keep sticking to their philosophy because they believe there are some clients interested in their designs. Howard Roark gives me the courage to follow what I believe. And I’m happy to have a small group of like-minded clients reading Singular Diligence.

Talk to Quan about What He Hates and Likes about Singular Diligence

Check out Singular Diligence