It’s kind of like mistaking the accelerator for the brake. Neither the gas pedal nor the brake are dangerous on their own. But making the mistake of thinking one is the other can have deadly consequences for the driver.
In the same way, making the mistake of thinking a company that makes acquisitions is just like a company that doesn’t make acquisitions can have deadly consequences for the investor.
http://www.investorquestionspodcast.com/storage/investor-questions-podcast-episodes/IQP_0020_How_Do_You_Calculate_The_Intrinsic_Value_Of_A_Company_That_Makes_Acquisitions.mp3 There are two issues here. One is growth. The other is capital allocation.
Let’s start with growth. Personally, I’m with Benjamin Graham when it comes to growth. Ben Graham said a stock’s earnings growth was best viewed as a qualitative rather than a quantitative factor. I agree.
When I look at any stock, including a stock like Regis (NYSE:RGS), I want to buy that stock for a price that gives me a margin of safety. And I want that margin of safety to come from the stock’s current balance sheet and past free cash flow record. Growth is good. Other things being equal, a stock with more growth ahead of it is safer than a stock with less growth ahead of it. But other things are almost never equal.
What Ben Graham and I want to warn you against is using geometry. If it’s a choice between using arithmetic or using geometry, pick arithmetic. When you project an earnings growth rate into the future, you’re using geometry. A lot of the stock analysis you see in magazines like Barron’s is a combination of geometry and a Keynesian beauty contest.
It goes like this. First, you pick a stock. We’ll use Johnson & Johnson (NYSE:JNJ). It earned $4.45 a share in 2009. Over the last 10 years, J&J’s earnings growth rate was 10%. So, we multiply $4.45 times 1.1 for – let’s say – 5 years. That’s $4.45 times 1.1 raised to the 5th power. The answer is $7.17 a share. That’s what J&J will earn in 2014. We then pluck a price-to-earnings multiple out of thin air – 15 sounds good to me, how bout you? – and we multiply the future earnings per share by the future P/E ratio. The answer is $107.55. That’s what JNJ stock will sell for in 2014.
Do I really think that’s true? I have no idea. There are two big assumptions. One is how fast JNJ’s earnings will grow. The other is how much investors will be willing to pay for each dollar of JNJ’s earnings in the year 2014.
Like I said, you’re really starting with geometry and ending with a Keynesian beauty contest. It’s a shaky method. Not because of the logic. I actually think the logic is sound. In theory, it works perfectly. In practice, it’s not the kind of method I think a mere mortal should use to pick stocks.
Why not? First of all, I picked a really benign example. Your chances of predicting wildly inaccurate earnings for J&J a few years out are much lower than your chances of predicting wildly inaccurate earnings for most stocks. J&J is predictable. Most companies are not.
Second, I used J&J’s 10-year earnings growth rate. If you read most stock analysis on the web and in magazines like Barron’s, you’ll notice the author cites much shorter growth rates.
Why is that dangerous? Remember, how I said investors are doing geometry when they project earnings growth rates? I wasn’t kidding. Back in high school you probably learned some physics. And one of the first things you were taught was something called velocity. And your teacher probably started teaching you about velocity using the idea of speed.
We all know what speed is. And when we say speed, like growth, we mean an average rate of change over a certain amount of time. In everyday use, we tend to mean a pretty long amount of time. And we tend to be talking about something that might change a lot from moment to moment, like a car’s speed, but gets you from one place to another pretty much in line with the average speed you expected.
You know how fast you can drive on a highway. So if I tell you something is 140 miles away, and it’s all highway driving, and there isn’t going to be any traffic – well, you know about how long it’s going to take you.
Velocity is a little different. It works much better with simple stuff. Steady stuff. I’m not sure you want to use velocity to estimate how long a car trip is going to take. And I know you don’t want to use it to estimate where a stock’s earnings are going to be.
Why not? Because a stock’s earnings both grow and shrink. Sometimes the rate of change is positive and sometimes it’s negative. In other words, most stock’s earnings are not – at least over just a few years – going to curve ever upwards. Instead they look more like a wave. There’s a downward curve at one spot, and an upward curve in another spot.
And that’s where using short-term growth rates becomes incredibly dangerous. When you project earnings, you either draw a straight line or a curve. You don’t draw a squiggly line. And you don’t draw a wave. If you take a really long average growth rate over – say 10 or 20 years – you can kind of draw a line through that wave. But if you take a short-term average, you aren’t drawing a line using an average speed that comes from both positive and negative growth – instead you’re taking something that looks a lot more like a stock’s earnings velocity going into or coming out of a peak or valley. Basically, you’re drawing a tangent. All you’re doing is just touching the curve for a second. And that’s how long your earnings growth estimate will be accurate for: a second.
If you look at the kind of earnings analysts predict at different points in the business cycle, you’ll see I’m not too far off when I say that projecting earnings means drawing a straight line into the future that just barely kisses the curve we’re all experiencing at the moment.
A better method is to take the long-term average speed of earnings over 10 years or more and use that to draw a line that is higher than the valleys but lower than the peaks. Even that method is far from perfect. To the extent that it matches GDP growth, you’re probably not going to be too far off. But you can’t project faster than GDP growth forever. No company or industry can take a bigger slice of the economic pie forever. Eventually, it has to slow down to grow in line with the economy. And, usually, a once faster than GDP growth industry will one day become a slower than GDP growth industry. Some industries don’t grow. They decay.
And that’s what we’re talking about here. Even the most predictable companies are more like populations you study in biology than curves you draw in geometry. There are limits to their growth. There are times when they decay instead of growing.
So that’s why Ben Graham and I think a stock’s earnings growth should be seen as a qualitative factor instead of a quantitative factor. I don’t want you to put growth into numbers. I don’t want you to do any math with it. I just want you to think about whether the company and the industry will decay, stay stable, grow slower than GDP, grow in line with GDP, or grow faster than GDP over the next 10 years or so. And, finally, I also want you to have some very rough idea of what the distant future will look like starting 10 years out. In other words, if you think an industry will grow fast for 10 years and then start decaying, you need to keep that in mind. Investors try to see the future. They don’t have a crystal ball. But they do try to grab a parachute before the plane slams into the mountain. So, it’s never a good idea to assume you’ll have plenty of time to sell out of a decaying business.
Regis is a great example to work with, because it’s in a predictable industry. I think hair salons, which is what Regis owns, are about as predictable as any industry out there. And I think they’re predictable both in the short-run and the long-run.
You know I own shares of Birner Dental Management (NASDAQ:BDMS). Regis is a lot like Birner. It’s in a huge, highly fragmented industry. The industry is basically no-growth. The CEO of Regis claims the hair care industry grows 2% a year. If you look at estimates for the dental industry, real growth is not much faster than that. But, in both cases, we are talking about growth in line with population at the very least. In the United States, population growth is about 1%. Long-term, real GDP per capita growth is maybe 1.8%. I wouldn’t put it above 2%. And I wouldn’t expect dentistry or hair care to grow their slice of the economic pie. So, a combination of GDP per capita growth and population growth gives us an upper limit of 2.8% plus inflation for these two industries.
I’m not an economist. I’m not even an industry analyst. But I like those numbers. And I’m willing to invest with them in mind. So, we’re probably talking about growth of 1% to 3% before inflation. How much growth you have in each industry after inflation depends on pricing power. Hair salons and dentist offices are local businesses. Regis is by far the biggest player in American hair salons and it has less than 5% of the market. So, I’m thinking customers who go to those hair salons and dentist offices are going to see prices rise along with inflation.
Of course, we don’t really care about prices, we care about profits. But I want to stick to today’s question. The question is how to calculate the intrinsic value of a company that makes acquisitions. To do that, we only need to look at growth and capital allocation. The issue of profitability is exactly the same whether the company makes acquisitions or not. That’s especially true in a highly local, highly fragmented industry. Acquisitions themselves shouldn’t have big effects on profit margins.
So, I’m going to say that organic growth in the hair care business is a lot like organic growth in the dental business. Don’t expect more than 3% growth before inflation. But do expect more than 1% growth before inflation.
Now, when analyzing Regis, you want to ignore the company’s long-term growth in sales, EBITDA, free cash flow, earnings, etc. and instead substitute this 1% to 3% organic growth rate. But, again, I don’t want you to do any math here. I just want you to get comfortable with whether or not you think hair care is a growing, stable, or decaying industry. I think it’s a stable industry that will grow at least as fast as the population but no faster than GDP. Like Ben Graham, my advice is to take that word stable as a qualitative factor in your analysis and forget about doing any math about growth rates. With growth rates this low, getting future growth estimates right won’t dramatically change your intrinsic value estimate for Regis, and getting future growth rates wrong could be dangerous. So stick to what you know. And just tell yourself Regis is a stable business. It’s worth the cash flow you can see today. Don’t pay up for tomorrow’s cash flow. But also don’t penalize Regis like it’s some buggy whip industry. It isn’t. Hair salons will be around forever.
So the first thing to do when calculating the intrinsic value of a company that makes acquisitions, is to find the long-term growth rate apart from acquisitions. We did that for Regis. I think it’s in the 1% to 3% range. You can go back and look at Regis’s past annual reports to find the company’s organic growth rate. It should be low but positive.
Now that we’ve stripped out Regis’s growth from acquisitions, we need to look at the company’s capital allocation.
Free cash flow is cash that could be paid out to shareholders. It is not, however, cash that actually will be paid out to shareholders. In almost all cases, dividends are less than free cash flow. Companies keep some of their free cash flow to make acquisitions, buy back stock, and pay down debt.
None of those things are good or bad, better or worse, in and of themselves. Sometimes acquisitions are a smart use of cash. A lot of times they are a dumb use of cash. Sometimes share buybacks make sense. Sometimes they don’t. Paying down debt works the same way. It sounds great. But I’ve seen companies pay down low-yield debt instead of buying back high free cash flow yield stock. That’s actually a bad idea. In the long-run, shareholders will do better if you leave the debt with a 6% interest rate alone and buy back the stock with the 13% free cash flow yield.
Let’s look at how Regis has spent its free cash flow in the past and how it might spend it in the future.
Full disclosure: I owned shares of Regis very briefly in early 2009. I sold them for a lot of reasons. Mostly because other stocks were getting cheaper faster. The best reason to sell a stock is because you’ve got something even better to buy. That’s almost always why I sell stocks. But there was another issue with Regis.
I didn’t like how the company managed its finances. Regis got itself in some debt trouble during the 2008 financial crisis. Nothing new there. Plenty of companies, including General Electric (NYSE:GE) and Goldman Sachs (NYSE:GS), got themselves in trouble during the credit crisis. Those two needed Warren Buffett to bail them out. Aren’t I being too tough on Regis? It only made the same mistake everybody else made, right?
Maybe. I owned the stock. And I didn’t like being diluted. I didn’t like having my piece of the pie shrink. No one does. And maybe I do hold a grudge.
But here’s my thinking on the issue. Regis wasn’t a bank. Regis wasn’t in a business that was credit sensitive. The salons themselves held up better than most businesses. Yes, 2009 was the worst year in Regis’s 84-year history in terms of same-store sales. But, you know what? At its worst, we’re only talking about 5% fewer people in the salons. A business should be strong enough to survive a 5% decline in customers coming through the door. Usually, having 5% fewer customer doesn’t mean you need a massive cash infusion.
But Regis did. Why? Because Regis spent its cash badly. It didn’t space out its debt. And it didn’t use long-term debt. Regis basically took whatever looked like the cheapest option available at the time and that’s what it used to finance itself. That was dumb.
I don’t expect the management at a company like Regis to foresee a credit crisis. But I do expect them to run the company’s finances in a way that doesn’t leave a super predictable business exposed to super unpredictable risks. They didn’t do that. And shareholders suffered.
Regis issued new shares of stock and convertible debt that gave away more than 20% of the company. If you do the math using the kind of intrinsic value calculations I like to use, you’ll see that Regis took more than $8 a share from existing shareholders and gave that value to the new stock and debt holders. Basically, they sold stock that would have been worth more than $30 in normal times and sold it for less than half that amount.
It was horrible. And it was unnecessary. About half the shares issued were shares the company had itself bought back around $30 a share in the years leading up to 2009. That means the company bought back stock at $30 a share only to re-sell those shares at less than $13 a share.
My point is that Regis financed itself in a riskier way than it needed to. The fact that the company was able to raise hundreds of millions of dollars by simply issuing new stock is not an endorsement of management. It’s actually a sign that management screwed up badly. When you can get out of a big debt problem by just issuing stock, that means the core business is healthy but the balance sheet is not. And it means the stock is cheap.
All of that was true of Regis in 2008 and 2009. If you look back over the last 10 years – or even the last 3 years – you’ll see that much of the debt problem wasn’t caused by acquiring too many hair salons. It was actually caused by buying back stock and paying a small dividend. There was no need to do either of those things with a balance sheet that was far from perfect and a voracious appetite for more acquisitions.
What bothers me most about all this is the 2009 letter to shareholders. The CEO, who I have to admit has always been open and honest with shareholders about Regis’s business performance, wrote a letter that I think was far from honest about the company’s financial performance.
He didn’t lie. He explains the transaction in the letter. But he does not apologize for it. He does not say it was a necessary evil. He talks about it like it was a good thing. It was a very bad thing. And, personally, I think it was an avoidable thing. It was an unforced error. And it would have been nice if the CEO admitted the mistake. It would have been nice if he was upfront about the huge amount of value destruction caused by bad timing and bad balance sheet management. That would’ve been nice.
It didn’t happen. But then again 2008 probably won’t happen again anytime soon. New buyers of Regis stock won’t suffer from past mistakes. All they need to know is what Regis will do in the future.
So is Regis a good buy? The stock’s vital signs are decent. It has a Z-Score of 3.02, an F-Score of 4, a 10-year free cash flow margin variation coefficient of 0.40, and a 10-year real free cash flow yield of 13%.
A 13% free cash flow yield is great for any stock. And this is a super predictable stock. Reading about Regis in preparation for this podcast reminded me of everything I liked about the business. I still like the business. And I love the 13% free cash flow yield.
It’s hard to see how buying Regis under $15 a share and holding on to it for any real length of time will end worse for you than buying the S&P 500. This is one cheap stock. And it’s my favorite kind of cheap stock: a predictable one.
The big questions here are free cash flow allocation and balance sheet management. There are plenty of opportunities for good free cash flow allocation. Regis can buy more and more salons for decades and still not run out of salons to buy.
I like the stock. If the balance sheet is managed conservatively, Regis is a super low risk investment. Right now, the Z-Score is marginal, and that’s without including big operating leases. But, I don’t think Regis is anywhere near as risky as most debt measures imply, because the business is very predictable. The issue I have is not with the amount of debt, but with the way I think that debt be handled in the future.
Still, I think Regis is an interesting opportunity. It’s definitely a stock worth analyzing. It would be on my list of top investment candidates. But, after the experience 2009, I doubt it’ll be in my portfolio anytime soon.
Regardless, it’s a great example of how to calculate the intrinsic value of a company that makes acquisitions. Basically, you just strip out growth from acquisitions and you focus on how free cash flow is spent. In this case, multiplying the 10-year average real free cash flow per share by the long-term average Shiller P/E of 14.88 gives you $28.11 a share.
That sounds right to me. But remember this is a leveraged stock. There’s a lot of financial leverage here. On the other hand, there’s almost no operational leverage. The company can handle a drop in sales well. And those drops are pretty rare.
In fact, there are stocks with almost no debt that are every bit as risky as Regis is when it’s carrying a fair amount of debt. But, personally, I’m still unsure about the way Regis handles its balance sheet. There’s no need for Regis to fly as close to the sun as it seems to want to.
But there’s also no reason for us to obsess about another credit crisis just because one just happened. The time to worry about extraordinary events is before they happen. Not after.
Regis is a cheap stock. I don’t own it anymore. But you might want to.
That’s all for today’s show. If you have an investing question you want answered call 1-800-604-1929 and leave me a voice mail. That’s 1-800-604-1929.
Thanks for listening.