Someone who reads my articles sent me this email::
My question is regarding the qualitative aspects of investing. How do you get comfortable with your understanding of a business? What sources do you use for information? How much research do you do before you decide you don’t want to invest in a business?
Start with the stats. A lot of people draw too sharp a line between the qualitative and quantitative aspects of investing. Accounting is the language of business. Look for stats that describe. Don’t use stats to predict. But dig into them and see what they tell you.
That’s what I always do. I always start with the past results. I always go back as far as I can with the stock. If a company has been public since the early 1990s, you’ll be able to find about 17 years of data at EDGAR. That’s because everybody was reporting electronically by around 1995.
So I go to EDGAR. And I create an Excel spreadsheet. And I enter all the key info for the last 17 years. That’s how I start. I’m looking for obvious things. I’m using a common sense approach.
If they have unit volume data, same store sales, market share, etc. I enter that for all 17 years too. Everybody will have basic financial info. Income statement stuff. Balance sheet stuff.
There will always be enough data to calculate operating margins, returns on tangible capital, etc.
I focus on the variation in these numbers. I’m looking for something extraordinary. The one thing I have zero interest in is an ordinary stock. I need an extraordinarily cheap stock. Or an extraordinarily good business. So those are the first things I look for.
The best sign of a good business is stable margins or stable returns on capital. I’m not interested in the trend. Certainly not the short-term trend. Maybe the trend over 15 years. But not over the last 3 years or 5 years. That could be cyclical.
Why do I care about variation? Why is a stable business a better business?
It isn’t. Stability is not necessarily a desired outcome. I’m not asking the company to aim for stability. What I am looking for is a symptom. There is a disease out there called competition. It is the thing long-term investors fear most. And the symptom of competition is the ability of outside forces to move key metrics.
There are a lot of metrics that matter. It depends on the kind of business you are looking at. Unit costs are key in some businesses. Same store sales could be a key metric. The combined ratio can be a key metric. Market share can be a key metric. It depends on the industry. Whatever those key metrics are – you want to gather them.
Then you want to read the 10-K and take notes. The best way to do this is have a highlighter, pen and paper with you. You will also want a calculator. Read a page and highlight anything worth remembering.
At the end of each page, put down your highlighter and pick up the pen. Look back at what you highlighted. Write questions in the margins. Write notes on your blank paper. Do calculations.
Let’s see what that would look like. I just read the Western Union (NYSE:WU) 10-K. Here are some of my notes.
This is a network business. Half of the company’s sales come from its top 40 agents. Western Union’s core business is consumer-to-consumer money transfer.
I’m only talking about the “consumer-to-consumer” business here. That’s a bit of a misnomer. Consumption is not involved. It just means person-to-person transfers. Not business to business transfers. Or household to business transfer.
It’s exactly what it sounds like. It’s migrants sending money home. That’s the key business. It’s not the only segment. But it’s the segment that matters most. So I focus in on that right away.
So, just by focusing on the “consumer to consumer” business we are talking about 84% of the company’s revenue. This is from the 10-K. There is a small domestic part to that business. But it’s small. About 85% of that 84% – or 72% of total revenue – is definitely consumer to consumer cross border transfers.
Most of that business is done through Western Union’s top 40 agents. These 40 agents – they can be banks, post offices, etc. – provide about 50% of Western Union’s revenue.
These are mature relationships. The average “top 40” agent has been with Western Union for 16 years now. That’s not surprising. Western Union agents make very little money in their first few years. Revenue per location tends to increase with age. Only about two out of three agents are truly active. About one out of three – especially in places like Asia – aren’t actually doing any money transfer business despite the Western Union sign.
Overall, the number of agents is misleading relative to the quality. Good agents are important. The volume done by top agents versus the bulk of mediocre ones is very big. Keep this in mind. Getting and keeping good agents is key.
So we’re painting a picture of the business using the 10-K. For Western Union, the key parts of the business – outside of the company – that we have to think about are agents and customers.
We’ve already talked about agents. The one thing I didn’t mention – but you probably guessed already – is that agents are paid by commissions. These commissions vary completely with revenue.
They aren’t a fixed cost at all. That’s why most of Western Union’s cost structure is variable. Because “cost of services” is mostly commissions. The rest is Selling, General, and Administrative Costs.
Now let’s talk about the actual customer. The customer is someone who sends money. To a very large extent, they are presenting cash at a physical Western Union location. About 40% of volume comes from customers who use a Western Union loyalty card.
So we have a picture of a typical agent, a typical customer, and a typical transaction.
Now, let’s talk about the product economics. How does Western Union make money off this service?
There are two ways. One, is through foreign exchange. Western Union deals – for customers – in around 140 currencies. Putting customers aside – and looking just at agents and expenses – the business is actually done entirely in U.S. dollars and euros. So exposure to foreign currencies is caused by the gap in time for money transfer. It’s a matter of days. But it’s obviously continual.
So currency fluctuations are unimportant to Western Union. They move results a percent or two one way or the other – hardly more than at a multinational consumer products company.
So why does foreign exchange matter to Western Union?
Because they use it to make money. They – like most financial institutions – take advantage of retail customers by charging a much worse exchange rate than they can get currencies at for themselves. This provides a bit under 25% of their revenue from customers.
The other 75% comes from a transaction fee. The transaction fee is based on the principal amount. However, lower principal transactions – when people make small transfers – cost those customers more as a percent of the amount transferred than big money transfers.
This is fairly typical stuff. It reminds me of banks, PayPal, etc. However, it’s worth mentioning the speed of most Western Union money transfers. Retail customers use banks for slow money transfer. They write checks. That’s a big business. And it’s domestic. And it’s very, very slow. Things like Fedwire are not a big part of how retail customers use their bank.
Western Union is different. Does it need to be? That’s another question. Most of these transfers are done super fast relative to other ways retail customers move money. You can pick up the money – in person – in hours rather than days or weeks (in the case of a mailed check).
What else is a key takeaway from the 10-K? What else is unique about this business?
Well, fixed assets are basically nil. Western Union’s balance sheet is weird. It’s smaller than it looks. Settlement assets and obligations exactly offset. These are quick money transfers. Not like a loan book. Accountants won’t cancel these items out. But investors probably should. That’s $3 billion on each side of the ledger.
They have a lot of debt. But it’s partly offset by cash. Leaving a manageable – and well spread out, partially fixed interest rate – debt load compared to operating income.
I had expected virtually all the cash to be outside the U.S. because that is where Western Union earns so much money – and because it has a low tax rate suggesting (this was verified by the 10-K) that it’s trying not to repatriate foreign cash. But they have more cash in the U.S. than I thought they would.
There are a ton of lawsuits. And a ton of settlements. This was expected. Western Union is a financial institution serving poor customers. Companies like that get sued a lot. If you’re familiar with pawn shops, payday lending, subprime, etc. – that’s a lot of what you see in the legal section here.
Some of it isn’t. There’s a tax issue. And there’s compliance type stuff – not consumer protection related. This brings up the issue of regulation. Is it good or bad for Western Union?
I think – and this is a very early guess based on reading a 10-K and just starting my research in the last few weeks – that regulation comes in two forms. One is good for Western Union. One is very, very bad.
Anything that has to do with how money and info is handled – rather than how customers are treated – is good. It creates a barrier to entry. Anti-money laundering rules are good for Western Union. They will violate them sometimes. But new entrants would have an even harder time. It’s a headache that helps keep others away. It’ll cost them in fines. But it’ll cut down on competition.
Consumer protection is very bad for Western Union. The marginal cost of a money transfer is low. What Western Union charges is high. Consumer advocates will focus on that. And say Western Union is preying on customers with absurdly high prices. Operating margins are fat in this business. Western Union’s operating margin is about 25% right now. And that’s coming from the consumer to consumer money transfer business. Which consumer protection would focus on.
That’s about it in terms of my notes. It probably took an hour to read the 10-K – technically, it’s about 160 pages, but they’re not all equally important.
There were a few other items. Kind of miscellaneous. Like and underfunded pension plan with too aggressive assumptions about future returns. That’s typical. If a company has a pension plan – it’s always too aggressive in accounting for the return on the plan.
I also jotted down some notes on things like the ages of top management. It’s a very small range. Everyone is early 40s to early 50s. This jives with the history of Western Union. It’s a professional management team. Nothing like the net-nets and so on I usually write about. Very typical for a big, public company.
That’s about it in terms of the 10-K. There is more in the proxy statement. And the 10-Q mostly duplicates what is in the 10-K.
So what other sources do I use?
Western Union’s public competitor is MoneyGram (NASDAQ:MGI). I spend a lot of time studying competitors. Basically, equal time to the company I’m interested in. I’m not interested in MoneyGram stock. So I’m only reading their SEC reports for info I’ll use to analyze Western Union.
Long-term data is unavailable for Western Union. There is a scattering of references in old articles, other reports, etc. that sketch out what the company looked like. But it’s only been a clear, separate entity for about six years.
This is one of those cases where the business has been around a lot longer than the corporation in its current form. So there’s no perfect year-by-year Value Line type data sheet you can create. In that sense, you can only go back to 2006.
I’ve already listened to the last few earnings calls. And I’ve read the company’s presentations. Western Union puts all that on their website.
So this is a pretty typical example of how I read a 10-K. What am I looking for?
The two keys are some measure of price to normal earnings – really, free cash flow – and a sense of how durable that annual free cash flow is.
A lot of people worry about growth. I don’t. For me, the most important part of an investment is how close it comes to being a perpetual annuity. So I go into something like Western Union wondering – is this a 5%, 10%, or 15% free cash flow yield? And what will free cash flow be in 2015 and 2018 and so on.
Basically, I would like to find a stock where the free cash flow yield is 10% - that is, I’m getting $1 in free cash flow for every $10 I pay for the stock – where I think free cash flow will be just as high in 10 years as it is today.
A key part of that analysis is the need for reinvestment in the business. At Western Union – it’s very, very low. You don’t need a lot of capital to grow this business.
In fact, the company can use free cash flow to buy back stock and thereby grow EPS faster than net income.
That’s what I’m hoping. One of the things that drew me to Western Union was that I glanced at some financial data and thought: “They should be able to buy back a lot of stock over the next few years.”
That’s a huge thing for me. I only invest in companies with falling – not rising – share counts.
So that’s my research process. Or at least the first step. It’s how I read a 10-K. I’d read it differently if I was looking at a net-net. Or some other balance sheet bargain. But for a stock where I am buying earnings and free cash flow – that’s how I’d do it.
Basically, you just read the 10-K and try to sketch out the business. You look for clues about the company’s competitive position, product economics, etc.
As always, the most important thing is getting a clear idea of customer behavior and management behavior. How do customers – and in Western Union’s case, agents – decide who to go with? And what is management going to do with future free cash flow?
Those are always the big questions. How much free cash flow is there? How reliable is it? And how long will it keep flowing?
You asked how quickly I decide not to invest in a company. It can be pretty fast. I decided I was interested in Western Union and wasn’t interested in Moneygram in a matter of minutes.
Some of you may remember Moneygram blew up – from bad investments – during the crisis. That’s true. But that wasn’t my issue.
What happened is that I did a quick check of the two company’s competitive positions. And I decided that Moneygram’s ability to attract the very best agents was not as good as Western Union’s. And that its brand name was vastly inferior to Western Union’s.
A “good brand” is kind of a vague concept. It’s made up of positioning, recognition, loyalty, etc.
The positioning in this business is clear. Moneygram uses direct price comparisons with Western Union – they always undercut on price – in their ads. They position themselves as the cheaper No. 2 offering the same exact service.
That’s not what I was looking at. I was looking at name recognition. And Western Union’s name recognition is astronomical. Moneygram can do a lot of things. But they can never close that gap.
For that reason, I immediately crossed off Moneygram as a possible buy. This is a fragmented industry. So they may do wonderfully for a long time. But I didn’t want to chance it. Western Union has the inside track.
Ask Geoff about His Research Process
Read Geoff’s Other Articles