• Page Title:
• Page URL:

This page has been successfully added into your Bookmark.

421 followers

Follow

Someone who reads my articles asked me this question :

Geoff,

Dun & Bradstreet Corp (DNB ) is a great buy in my opinion. I have been considering it myself since it dropped to the mid-$60 range last week. I have one concern about it and this holds true for a few other stocks that have this same characteristic. Several very successful companies have been buying back stock at a rapid pace and this has caused shareholders equity to go negative. Direct TV (DTV ). AutoZone (AZO ) and Dun and Bradstreet (DNB ) are examples. All have great track records for business and stock price growth long term so this appears to be a strategic decision to finance through debt rather than equity. My question is at what point can an investor judge that this practice is no longer creating value but adding risk to the investment? There can obviously be too much leverage. However when money is cheap to borrow and ROIC is high this seems like a great way to create value for shareholders, to just borrow at 3% and buy stock with an ROIC over 20%, but how much is too much?

Thank you,

Jeff

It’ll take me a while to answer. And I’m afraid my answer may really go off on some tangents here.

So, I want to make a few key points right away:

· Negative equity itself is meaningless (could be good or bad).

· Operating liabilities and financial liabilities should be analyzed separately.

· You will often have to restate the value of assets from book value if you want the balance sheet to reflect reality.

· In special cases – like with pensions – you may have to restate liabilities as well.

· Liquidating safety and operating safety are two different things.

· Compare net financial obligations to EBITDA.

· Compare net financial obligations to free cash flow – think of borrowed money as the price of time (a good business can always wait a few years and do the same things without any debt, ask yourself which you’d rather they do – borrow money or spend time).

· Look at EV/EBITDA – not just the P/E ratio.

Finally, remember you can always learn from someone who invests in debt.

Stock investors can learn a lot from people who analyze debt. Here is a good blog to give you a taste of what I mean.

Reading about investing in debt – and especially about companies in bankruptcies – can give you a new perspective on situations like these.

I just wrote an article about how free cash flow isn’t everything. Well neither are earnings. And neither is book value.

If you look at the stocks in my portfolio, you’ll find I must value free cash flow very highly because as a group they tend to:

· Always have positive free cash flow

· Have unusually high ratios of free cash flow to reported earnings

· Buy back shares

· Pay dividends

· And still have excess cash

Dun & Bradstreet sure doesn’t meet the last criterion (it has a lot more debt than cash). But it checks the other boxes. This is something both my “wide moat” investments and my “net-nets” have in common. Free cash flow.

And yet, I wrote a whole article about how free cash flow isn’t everything.

That’s because it’s really looking at the space between the three key statements:

· Balance sheet

· Income statement

· Cash flow statement

Where you do your most important work.

Investing isn’t about the balance sheet or the income statement or the cash flow statement. It’s about how the balance sheet, income statement and cash flow statement interact through time.

Never be myopically focused on one financial statement or myopically focused on one time period.

Each statement reveals just one aspect of the same object: a business. And it does it for just one time period (in fact, the balance sheet does it for just one moment in time).

As an investor, you need to be able to see all aspects of the business in motion.

Because we are seeing just one aspect of a business when we see something like negative equity – that fact alone means almost nothing.

It would be like if we were analyzing football players and we knew somebody weighed 300 pounds. Is that good or bad?

For a quarterback?

Bad.

On the line?

If he’s good enough in other ways – he could weigh even more and nobody would mind.

If we were drafting a football player we’d want to know the combination of age, size, weight, agility, skill, personality, etc. Just knowing weight isn’t going to help without having more context of how that one aspect relates to the whole player.

Same thing with a business. Same thing with an investment.

So negative shareholder’s equity alone doesn’t matter. In fact, like you said, it can be something you find with stocks that perform perfectly fine over time.

Now, about Dun & Bradstreet.

I wish I had bought more Dun & Bradstreet (DNB ). On the morning of the big drop, I only had 6% of my portfolio in cash and ready to buy. I’m doing a bit of selling of something else – over in Japan – right now. It’s a cheap stock. And will probably perform wonderfully for whoever buys it from me. But the risk of catastrophic loss rose since I first bought the stock. The risk/reward may still be good. But the reliability is not as high. That’s usually my cue to go. So I would be selling that Japanese stock anyway. Even if DNB hadn’t dropped.

But I do hope it sells a little faster – now that I have someplace to put the cash.

I hate having only 6% of my portfolio in a position. I have a pretty strong dislike – almost a rule against – any position less than 10% in my portfolio. I have found I do not make good decisions when I have to juggle 10 or more opportunities in my head at once.

That shows you how much I like DNB. For most stocks, I would wait until I knew I had enough cash to put more than 10% into the stock. For DNB, I immediately used all the cash I had available. Even before I worried about how I could get more. If the price of DNB gets away from me – and I get stuck with just a 6% position – I will be disappointed. But I won’t sell the position. I’ll hold it. Probably for a long time. You don’t get many opportunities to buy a business with the super wide moat and essentially non-existent tangible investment requirements of something like DNB very often.

There are some companies where – if I feel the business is still as I remember it, economically – I will buy the stock whenever it gets to an acceptable owner earnings yield. There is not much more to it than that. “Acceptable” in my book is better than my next best alternative or 10% a year – whichever is higher. If nothing seems priced to return about 10% a year, I hold cash.

I don’t believe in market timing. But I don’t believe in taking a risk where I think if everything goes perfectly the upside is still going to be in the single digits.

Well, after the big drop – I thought DNB was just about exactly priced to deliver 10% a year. I don’t mean that I know what the stock will do. But I think the normal economic earnings of the business – after they pay interest and all that – that could be passed on to owners will be about 10 cents a year for every $1 I paid for the stock last week.

Plenty of other stocks are priced in a similar way. But they are not reliable in the same way I think DNB is. Trading at a P/E of 10 is not the same thing as being priced like a 10% perpetual bond.

In most cases, it is very different. In DNB’s case I don’t think that’s true.

That was the rationale for the purchase.

Now, on to your concern. The share buybacks. And the negative equity.

So, negative equity alone has no meaning. It’s a non-issue.

I buy stocks all the time – most stocks I buy in fact – that have positive tangible equity in excess of my purchase price (that is, they trade below tangible book value). But I also buy stocks with negative book value. I owned IMS Health before it was bought out. They had the same practice as DNB. They borrowed and bought back stock. Year after year.

Different people have different ways of measuring how much debt is too much debt. This is a separate issue from whether the debt is being used productively.

Obviously, if you have to borrow at 9% and you can only earn 6% on what you borrow – that’s not a recipe for success. So, in that case, even if you are borrowing a “safe” amount of debt – you may still be doing wrong for shareholders.

But, here we are talking about companies that tend to use borrowed money to buy back shares – not buy equipment, develop land, dig for gold, etc. So the calculation should be a bit easier.

If the company is able to grow at least as fast as inflation and the stock trades at a price to owner earnings of say 10 – and you think it’s worth every penny – then obviously borrowing at anything less than 10% a year should theoretically be fine if all of the money borrowed is used to buy back stock.

Generally, I’d want to make sure that a company that is buying back stock while simultaneously owing money is always borrowing at – hopefully, longer fixed rates – that are lower than the return I’d expect on the stock if the company stopped growing today and never started growing again.

That would be my test. If a company can pass that test – it can go ahead and borrow and buy back stock. As long as the level of debt is also safe. We’ll deal with that issue in a minute.

So, if I think DNB stock could return close to 10% a year even if it was a truly no-growth company from this moment forward – then it’s okay that they borrow money at a much lower rate.

How much lower?

In their 10-K, DNB says that in November 2010, they borrowed $300 million at 2.9%. The $300 million is due in November 2015. So they borrowed for five years at under 3%. Buying back stock will return more than 3%. That was true even when the stock was a lot more expensive.

They also have an $800 million credit facility. They owed $260 million on that facility at the end of last year. They currently pay 1.6% a year in interest on that $260 million.

Again, 1.6% a year is much lower than the return DNB’s shareholders can expect from share buybacks.

So, borrowing money and buying back stock is okay from a return on investment perspective.

What about a safety perspective? Is the amount of debt DNB is carrying safe? Can they handle it?

Here, we don’t care that they are buying back stock. And as strange as it sounds – we don’t care that they have negative equity.

I know I write about net-nets all the time. I’m the guy who writes GuruFocus’ Ben Graham: Net-Net Newsletter. By definition, a net-net trades below book value. So you’d think I’d be a big believer in the importance of book value.

I’m not. Book value alone means nothing. It can hint at something big though.

Tangible book value is a useful screening tool. So is EV/EBITDA. Neither measure is perfect. They are more useful when you are soaring over the entire market trying to spot bargains. They are less useful when you are trying to analyze specific companies. If an entire country’s stock market has a low price-to-tangible book ratio or low EV/EBITDA this is very important info to know. In fact, it’s decisive. You can buy indexes on that knowledge alone. But you probably shouldn’t buy specific companies on that knowledge alone.

Ultimately, things like:

· Liquidation Value

· Market Value

· Replacement Value

· And Owner Earnings

Matter more. These things trump:

· Book Value

· EV/EBITDA

But you’ve got to calculate them yourself. You’ve got to move beyond being a record keeper – an accountant – and become an appraiser.

We’ve talked about this kind of thing before. DreamWorks Animation (DWA ) carries a library of animated movies on its books. For about $13.5 million a movie.

You can find all this for yourself by reading the company’s balance sheet and the notes to its financial statements in its 10-K. I can’t stress this enough. You always have to read the notes. In many ways, the notes are the 10-K.

If you could separate DreamWorks’ employees, property, technology, management and production pipeline from the movies they’ve already made – and their rights in those franchises – would you be willing to pay more than $14 million a movie for that intellectual property stub?

Some (old movies, and a couple flops) are carried at zero. A couple new ones are carried for a lot. All of them together are carried for about $310 million.

Would you pay $310 million for the movies DreamWorks has already released plus the rights to keep making movies in those franchises?

If so, then that balance sheet item is probably worth even more than $310 million. And DreamWorks’ book value – as intangible and full of intellectual property as it may be – is actually understated.

If not, then that balance sheet item is worth less than $310 million. And DreamWorks’ book value is overstated.

This is because the way DreamWorks treats their other inventory – the stuff that hasn’t been released yet – is pretty similar to how the accounting works at other companies in other industries (they make it and as they do – they carry it on their books at their cost). Once a movie is released, this gets trickier though. Because DreamWorks starts amortizing the movie at what may or may not be an economically accurate way of recording long-term revenue generating potential and residual value.

And so you have an asset that might not be carried on the books at what it could be sold for. And yet it might be quite possible to sell that asset.

You have a similar situation with land. Sometimes companies accumulate land over the years at prices that do not reflect what that land could be sold for. G enerally A ccepted A ccounting P rinciples (GAAP) does not allow companies to mark up the value of this land over time. This is particularly important to note when changes in the business make the land less integral to the business than it once was. In other words, some companies end up with valuable land they could sell without radically changing how they run their business day-to-day.

And then you have the very tricky concepts of receivables and inventory. They are good in liquidation – yes. You can borrow against them – yes.

But, generally, they are not a very good asset to own because they are utterly integral to the business and they need to be constantly replenished – essentially they become an obligation. You’d rather have less working capital than more working capital if you could.

So some assets on the balance sheet matter a lot more than other assets. And their book value may not reflect their market value.

The assets that matter most are usually:

· Cash

· Investments

· Land

· Intellectual Property

· Tax Savings

· Legal Claims

If you have these things, you can support less debt. If you don’t have these things, you can’t support debt – except to the extent you are generating cash flow from your business.

Cash flow from the business is always best. But if you’re focused on a static snapshot of a business – like a balance sheet – it helps to restate the cash, investments, land, intellectual property, etc. to reflect what they could be sold for.

Things that can be sold can sometimes be borrowed against. And selling things can help save shareholders when there is debt and no cash flow to pay for it.

But that is not a good situation. Cash flow protection is much better than asset protection.

Strong, reliable free cash flow is usually a surer sign of a company’s safety than anything you’ll find on the balance sheet.

But why those specific assets?

Why not include inventory, receivables, etc. Why talk about land – and stuff you can sue over?

These assets matter most because they are in some sense separable from the operating business itself. You convert these assets into other things. There are different ways of doing it. Not all of those things can be sold. But if you want to exit a certain line of business – you can usually keep those things.

They become corporate assets more than business assets.

Inventory and receivables are different. They are important. But they are most important as an acid test of cheapness and overcapitalization.

A net-net is almost always cheap and overcapitalized.

That’s why you screen for them. But, like I’ve said before, don’t fixate on a net-net’s assets. Just use those assets to prove the company is cheap. Then pivot and start analyzing the operating business – its profitability, reliability, future prospects, etc.

When it comes to an operating business – don’t think of assets as assets. Assets in a continuing business are not necessarily good. Liabilities in a continuing business are not necessarily bad.

In liquidation, assets are good. And in liquidation, liabilities are bad.

But we’re not talking about liquidation.

Liquidation should not be your first line of defense.

So, when deciding whether a company like Dun & Bradstreet, DirecTV, AutoZone, etc. is carrying a safe amount of debt – whether the balance sheet shown at book value’s verdict of negative net worth is economically accurate – you want to break the company down into:

· Cash

· Investments

· Land

· Intellectual Property

· Tax Savings

· Legal Claims

· Owner Earnings

While the right measure to use is owner earnings, I’m going to talk about these stocks you mentioned in terms of EBITDA. It’s a number we can all agree on. Yes. It is too generous. If people use EBITDA like it means EPS – they are trying to fool you. But EBITDA is not evil. It is a tool. As useful for analysts as for promoters.

We’ll try to use it responsibly here.

EBITDA saves us from debating the exact amount of maintenance cap-ex, working capital changes, etc. that would be needed to support a no-growth DNB, DTV, AZN, etc.

How much is EBITDA worth?

If you had to capitalize EBITDA like it was the rent on an apartment building to figure out the value of that apartment building – what number would you use?

It’s unlikely U.S. companies generate much more than 33 cents of EBITDA for every dollar of book value they have. I don’t have data on this. It may be a smidgeon higher at the moment. But that’s only because we live in odd times. Right now, returns on equity – however you measure them – are really high in the U.S.

So, a normal number would be even lower. And remember, U.S. stocks trade way above book value – so even if a company is generating 33 cents of EBITDA on its book value – investors only be getting more like 15 cents of EBITDA on every dollar of their cost in the stock.

That number may sound wonderfully high – but 15 cents is less than you think after you pay for physical depreciation (a real expense) and taxes (another real expense). Even without interest payments of any kind, you can easily go from 15 cents of EBITDA to about 7 cents of net income.

What I just described isn’t far from the current reality in the U.S.

Okay. So I think it’s fair to say that at a normal company you would need at least $3 of book value to generate $1 of EBITDA. Maybe more. But not less.

Here’s where I want to talk about asset-earnings equivalence.

I’ve mentioned before that assets generate earnings. And then earnings are used to buy (or build) assets. And then those assets create more earnings.

And so you have this cycle of investment. You have a stock of assets. You have a flow of earnings. You turn the flow into the stock. And you turn the stock into the flow.

Well, the truth is that when you have a flow of earnings on the income statement – and yes, the cash flow statement – but no assets (or very few assets) on the balance sheet, this doesn’t mean you have a negative economic net worth.

It means something different.

It usually means you have a special asset. An asset that is worth its flow. Not an asset that can be easily appraised perhaps. And certainly not an asset that can be easily compared to other assets.

This is not like owning a single family home on a street with 20 others.

This is like owning a highway rest stop. There is nothing else for 30 miles. Maybe you can build a restaurant here as cheaply as a restaurant anywhere. But the cash flows are going to be different. And so the market value of that rest stop location will be different regardless of what your original cost was – this is if and only if folks can’t put up a thousand other restaurants all around yours.

Well, Dun & Bradstreet is like that highway rest stop. DirecTV is like that. And Autozone is like that. To some extent. They all own special assets. Assets that are not separable from the operating business.

The business is the valuable asset. And it’s valuable in ways the balance sheet may not reflect.

Well, what if we just approximately applied this idea to $1 of EBITDA is similar to $3 of equity?

It’s a strange concept. But let’s see where it takes us.

DNB had EBITDA in 2011 of $506 million.

How much did it take to produce that $506 million in EBITDA?

The truth is that it took nothing. If you look at what I normally consider the core invested assets of an operating business:

· Receivables

· Inventory

· Property, plant, and equipment

And you net them against what I consider to be the core liabilities of an operating business:

· Accounts payable

· Accrued expenses

· Deferred revenue

You don’t get a positive number. You get a negative number.

What does this mean?

If you liquidated Dun & Bradstreet’s business – it would cost you money. There’s a reason that revenue is deferred – you’ve been pay, your customers haven’t been served – they won’t appreciate a sudden shut down. If you try to flip a switch and shut it down – you would not be able to take more cash out of it.

In fact, since you’ve been paid for services you haven’t provided – you’d actually have to put more money into DNB to shut it down. If you don’t provide the service – they’ll want their money back.

This is the opposite of a net-net. If you shut a net-net down – no one would need to inject more money into the company to liquidate it. Instead, it could be shut down and a surplus from the sale of inventory and the running off or sale of receivables could be paid out to owners.

It’s important to note that being in a strong, safe liquidating position does not necessarily mean you are in a strong, safe operating position.

Most net-nets have a higher risk of bankruptcy than Dun & Bradstreet. (Understatement of the century.) But all net-nets have a lower risk of failing to survive a forced liquidation than Dun & Bradstreet.

Does that matter?

Does it really matter to you how a stock like Dun & Bradstreet would fair in a liquidation?

I don’t think it does. In fact, if Dun & Bradstreet ended up in bankruptcy – what would happen to the operating business would be a much bigger concern than say the $118 million or so they have in cash. Yes – cash, receivables, etc. matter. But if you have a business producing $500 million in EBITDA, people want to preserve that asset. And a business that produces cash flow really is an asset. In fact, the operating business would be the key asset if a company like:

· Dun & Bradstreet

· DirecTV

· Or AutoZone

Couldn’t pay its creditors on time.

Let’s look back at that very theoretical mention of EBITDA I made. I said that an American public company with $1 of EBITDA might also have $3 of equity.

Now, equity is not the same as assets. Companies use leverage. But imagine for a moment what this would mean if a company with $500 of EBITDA followed the same sort of pattern as other companies.

Well, it would have at least $1.5 billion in net assets.

As I pointed out, Dun & Bradstreet – the core operations of the company, what we’ll call the business rather than the corporation – really doesn’t have any net assets. Its assets are less than its liabilities.

So we’ve got at least a $1.5 billion hole here.

In my book, that’s economic goodwill. Not accounting goodwill.

It’s the investment shareholders need to normally put into the business. And – in this case – it’s simply not there. The stock of invested assets is missing. But the cash flow is there. And the cash flow is what has value.

Now, if you take out all the assets and liabilities related to operations from DNB’s balance sheet you’re basically left with

Financial Assets

· Cash: $118 million

Financial Liabilities

· Debt: $843 million

· Pensions: $595 million

Let’s net them out. You get net financial liabilities of $1.32 billion.

So we’ve got an asset – this operating business with (what I think is conservatively calculated) economic goodwill of $1.5 billion – and we’ve got this $1.32 billion liability.

Is that different from buying a house for investment purposes and borrowing 88% of its appraised value?

I don’t think so. I think – if you need to look at it from an asset/liability perspective – that’s the right way to look at it.

Yes. It’s borrowing a lot of money. But it’s borrowing against the appraised value of the business – really what the market would pay for DNB’s cash flows. Not the book value of DNB’s business.

DNB has an operating business that would normally need $1.5 billion in net assets to support it. And it borrowed $1.32 billion.

This has very little to do with what the business is worth to a stockholder.

For that, you’d need to start talking about what EV/EBITDA is at DNB.

But you didn’t ask whether DNB was a good investment. You asked whether it was safe.

The real answer to whether companies like DNB, DTV and AZO are safe has to do with the kinds of measures people who look at debt worry about.

It usually involves EBITDA. Which I know is a dirty word among some value investors – and Charlie Munger in particular. But it isn’t easy to compare companies with different businesses and different financial structures.

For one thing, DNB – and some of these other companies – are already unusual from an operating perspective. And even EBITDA does not “solve” this problem.

Over the last 10 years, DNB has turned 68 cents of every dollar of EBITDA into free cash flow. This is rather remarkable when we consider that corporate taxes in the U.S. are 35%. Interest rates – while low – are still higher than zero. And DNB grew nominal sales by 3.5% a year over the last 10 years.

EBITDA should be reduced by:

· Interest

· Taxes

· Additions to property

· And additions to working capital

We’d expect that to cause free cash flow to come in closer to half of EBITDA than two-thirds in the kind of circumstances I described.

It didn’t at DNB because:

· Working capital is negative

· Capital spending needs are minimal

These aren’t financial aspects. They’re operational aspects of the business. In fact, they are core and usually quite difficult to change aspects of the business.

I usually find that working capital needs and capital spending needs are part of the DNA of a business. They are there from birth. They are – in broad strokes – something that’s very hard to change. You can improve discipline. But you can’t turn a railroad into an ad agency. Their property requirements are what they are. And they’ve always been that way.

An exceptionally cash-generative business is often an exceptionally cash generative business for reasons that have nothing to do with the current management team or their policies.

So even EBITDA doesn’t help us separate a company that converts one dollar of EBITDA into 50 cents of free cash flow over a decade from a company that converts one dollar of EBITDA into 68 cents of EBITDA over a decade.

So, no measure is perfect. When in doubt, creditors and shareholders would both prefer to see free cash flow come in higher. Free cash flow is the best protection.

But we’ll look at EBITDA because that is a customarily used measure – and it allows us to compare different companies without having me constantly talking about why free cash flow is high here but low here. EBITDA just causes fewer problems than either a net income or free cash flow based discussion would.

So, I mentioned that financial liabilities at DNB – basically debt and pensions – are $1.3 billion. And EBITDA is $500 million.

That means debt – we’ll lump pensions in with debt here – is 2.6 times EBITDA. Can you live with that?

Is debt of 2.6 times EBITDA okay?

That’s the question to ask. Not whether it’s okay to have negative equity. Negative equity itself is not a risk. Poor interest coverage is.

You can also measure the ability to repay debt by looking at free cash flow. For free cash flow – because of working capital swings – never use a one year number. Take a three-year average. In DNB’s case that three-year average is $325 million.

So, if DNB continued to produce free cash flow at the same rate – and used every penny of free cash flow to pay down its debt and fund its pension liabilities – it would take the company exactly four years to scrub its balance sheet spotless.

This is probably closer to the kind of number Warren Buffett would use. He’d probably say: “The company can pay everything off in four years.”

There is an additional problem with DNB. The pension plan. I said pensions were about $600 million. But that’s only the unfunded portion. The actual obligation is $1.7 billion. This is then reduced by the value of pension plan assets.

The calculation of the unfunded portion that appears on the balance sheet depends on assumptions DNB makes. Some of which are too aggressive:

· Expected Long-Term Return on Plan Assets: 8.25%

They are cutting it to 7.75% going forward. That is still too high.

The plan’s target allocation is:

· 55% stocks

· 43% bonds

· 2% real estate

We’ll call that:

· 55% stocks

· 45% bonds

Which is awfully close to 50/50 between stocks and bonds. So the math is pretty easy.

What would I say the expected return on a 50/50 stock and bond portfolio should be?

6.25%.

I think 50/50 stock and bond pension plans that are assuming more than 6.25% a year are assuming too much. And that obviously means they are understating their liabilities.

Of course, some people disagree with me. And some of them are a lot smarter than me. Berkshire Hathaway assumes a 7% a year return on their pension plans. At year end, stocks were 60% of Berkshire’s pension assets.

I think assumptions more aggressive than about:

· 8% for stocks

· 4% for bonds

Are too aggressive.

Berkshire may be right to use 7% for their pension plan return assumptions.

But, if it was up to me, I’d use 6%. Of course, nobody uses 6%.

The average expected return on pension assets is 7.8% at the 100 largest U.S. public companies. So, DNB is right in line with them with its 7.75% expectation.

But Dun & Bradstreet’s funded status is worse than most big U.S. companies. On average, companies have funded between 75% and 80% of their pension obligations.

Here is a comparison of expect return and actual returns for the 100 public companies with the largest defined benefit plans: