return on equity vs. return on capital: why this matters today more than usual


Let’s pretend we live in a world without taxes, just to make things simpler.

Year 1:  A start-up company raises $1,000,000 by issuing stock.  It uses the money to create a business that earns income of $100,000 a year.  Its return on equity = return on capital = 10%.  The firm reinvests all income into the business.

This is a pretty ho-hum business, returning 10% from operations.

Year 2:  Management then raises debt capital to supplement its equity by borrowing $900,000 from a bank at 5% interest.  It uses the extra funds to expand aggressively.

Let’s say it gets the same return as with its initial capital  Using the loan + retained profits from Year 1, it doubles the size of its business.  It earns $200,000 in income from operations  in Year 2 -$45,000 in interest expense = $155,000.

Its return on its total capital of $2 million, after deducting interest expense from operating income, drops, to 7.8%

Its return on equity of $1.1 million, however, rises,  to 14.1%.

The now financially-leveraged company posts 55% earnings growth, not 10%, and sports an above-average return on equity.

To the casual observer it now looks like a dynamo.   …but the transformation is all due to financial leverage.

Year 3:  Including income reinvested back into the business, the company now has $2,155,000 in capital, $1,255,000 of that in equity and $900,000 in debt.  It borrows another $900,000 on the same terms from its bank and puts that into the business.

The $3.055 million generates $305,500 in income from operations.  Interest on $1,800,000   @ 5% is $90,000.  Doing the subtraction, net earnings = $215,500.

Earnings growth is 39%+.  Return on equity is now 16%+.      Again, the difference between being the sleepy 10% grower and an apparent home run hitter is entirely due to management’s financial engineering.


What’s wrong with this picture?    In a bull market, nothing.  But the company has exposed itself to two financial risks if business slows.  Can it generate enough cash to pay the $90,000 in interest expense, which amounts to four months’ profits in good times but maybe ten months’ in bad?  Can the bank call part–or all–of the loan?  If so, how does our company get the money, which is the equivalent of nine years’ earnings?


stock buybacks:  more financial engineering

A second issue:  suppose the company employs another form of financial engineering and uses the money it borrows at the beginning of Year 3 to buy back stock rather than reinvest in the business.

Why do this?

…it doesn’t improve overall earnings, but boosts earnings per share.  Although framed in press releases as a “return” to shareholders, this also–one of my pet peeves–disguises/offsets the dilution of you and me as shareholders through the stock options management issues to itself.  (I’m not against stock options per se; I’m against the disguise.)  In this case, earnings are $215,500 before interest expense of $90,000.  Interest expense amounts to five months’ profits.  The loan principal is equal to 18 years’ earnings.


how/why does financial engineering like this happen?

When there’s lots of extra money sloshing around in the system, banks, the fixed income markets and companies do crazy things.  This was a potential worry several years ago.  Unfortunately, lacking understanding of how the economy or the financial system works, the Trump administration has made the problem worse through the tax and money policies it has pursued.  Instead of taking away the punch bowl, Trump has spiked it a lot more.


my take

For us as investors, the point of this post is to distinguish between companies that show high returns on equity because of the earning power of the company business (high returns on capital; these are keepers) from those where financial engineering is the main reason returns on equity are high (low returns on capital; riskier than they seem at first glance and likely to perform poorly in wobbly markets).







my take on Disney (DIS)

Post the Fox Studios acquisition, DIS remains in three businesses:  broadcasting, theme parks and movies.  December 2019 quarterly operating income from the three (ex direct to consumer) was roughly $5 billion.  Of that, just under half came from theme parks, a third came from broadcasting and the rest from movies.

All three business lines have their warts:

–ESPN, the largest part of broadcasting, has long since lost its allure as a growth vehicle; ABC is breakeven-ish; I think there’s some scope for boosting results from Fox.  Pencil in slow/no growth

–theme parks are (were?) booming, but they’re a highly business cycle sensitive business.  We’ll see that, I think, in March- and June-quarter results.  so even though it’s Disney we should apply a discount multiple to these earnings

–movies are traditionally a low multiple business because of the irregularity of new releases and their typical hit-and-miss nature.  DIS has been exceptionally good for a long time under Bob Iger (it was pretty awful before him), but this is still inherently a low multiple enterprise.


This is the main reason the stock spent years bouncing around the $110 level with flattish $10 billion, $6 a share, in earnings.  In the era of human analysts, it wasn’t just the earnings that held the company back.  It was also knowledge of the slow demise of ESPN, the (in hindsight too conservative) sense that theme parks were nearing a cyclical peak, and the idea that the company’s incredible movie hot streak might come to an end.

Then there’s the streaming business, where–to pluck a number out of the air, DIS is spending 10% of its operating income to develop.


A couple of months ago no price seemed too high for Wall Street to pay for Disney+, with the stock peaking at $153.    If we say $25 of the rise was due to streaming, at the top the market was valuing Disney+ at $40 billion+, or roughly a quarter of the value of Netflix.

The stock has fallen by about 40% since.

Now, hang onto your hat:

If we assume that the implied value of Disney+ has fallen in tandem with NFLX, it’s now valued at $32 billion.  This gives a residual value for the rest of DIS of about $140 billion.

That would imply a multiple of 13x current earnings–or, alternatively, an assumed 20% decline in profits.  The decline would likely be mostly (I’m assuming entirely) in the parks and movie divisions, implying that area’s income would fall by somewhere around 30%.

So–in this way of looking at things, we are assuming substantial success for Disney+ and a decline in the most cyclical businesses of roughly half what the market is assuming for companies like Marriott.  It would be cheaper to create a “synthetic” DIS out of NFLX + MAR shares, although what would be missing would be the Disney brand.

My conclusion:  Mickey and Minnie aren’t screaming “Buy me.”





business line analysis and sum-of-the-parts

This is mostly a reply to reader Alex’s comment on a post from early 2017 about Disney (DIS).

The most common, and in my opinion, most reliable method securities analysts use to project future earnings for multi-business companies is doing a separate analysis for each business line.  This effort is aided by an SEC requirement that such publicly traded companies disclose operating revenues and profits for each line of business it is in.

In the case of DIS, it’s involved in:  broadcast, including ESPN; movie production and distribution; theme parks and resorts; and sales of merchandise related to the other business lines.

There is plenty of comparative data–from trade associations, government bodies and the financials of publicly traded single-business firms–to help with the analysis.  And every company has, in theory at least, an investor relations department that answers questions put to it by investors. ( My experience since retiring as a money manager for institutional clients is that many backward-thinking well-established companies–DIS and Intel come to mind–can be distinctly unhelpful to their most important supporters, you and me.  (To be fair, I haven’t spoken with DIS’s IR people for several years, so they may be better now.))

Analysis consists in projecting revenues/ profits for each business line and using the results as the key to constructing a series of whole-company income statements–one each for this year, next year and the year after that.

The trickiest part is to decide how to value this earnings stream.  The ability to do this well either comes with experience or from having worked for a professional investor who’s willing to teach.


More tomorrow, or in a day or two if I don’t get my film editing homework done today.




Whole Foods (WFM) and Amazon (AMZN)

Most of the talk I’ve heard about AMZN’s acquisition of WFM revolves around the idea that AMZN plays a long game.  That is, the company is willing to forgo profits for an extended period in order to achieve market share objectives–which will ultimately lead to an earnings payoff.  After all, it took eight years to get its online business into the black.

What’s being lost in the discussion, I think, is the present state of WFM.  It’s not a particularly well-run company.  Analyst comments, which have surfaced publicly only after it became clear that WFM would be acquired, suggest the company has antiquated computer control systems.  It has waffled between emphasis on large stores and small.  We know that it needed a private equity bailout during the recent recession.    It has begun a down-market expansion through “365 by Whole Foods” stores; in every case I can think of, except for Tiffany, this has been a sure-fire recipe for destroying the upmarket main brand.

The easiest way to see management issue, I think, is to compare WFM with Kroger (KR), a well-run supermarket company.  Their accounting conventions aren’t precisely the same, but I don’t think that makes much difference for my point.   (Figures are taken from the most recent 10Qs.).  Here goes:

–gross margin:  KR = 19.7%;  WFM = 33.8%

–pretax margin:  KR = 1.2%;  WFM = 4%

–inventory turns/quarter:  KR = 5.8x; WFM = 7.7x.

What do these figures mean?

–WFM turns its inventories much faster than KR, which should give WFM a profit advantage

–WFM marks up the items it sells by an average of about 50% over its cost of goods;  the markup for KR is half that.

–the combination of faster turns and much higher markup should mean a wildly higher pre-tax margin for WFM

–however, 14 percentage points of margin advantage for WFM at the gross line almost completely evaporates into 2.8 percentage points at the pre-tax line.

–this means that WFM somehow loses 11.2 percentage points in margin between the arrival of goods in the store and their delivery to customers, despite the fact that stuff sells significantly faster at WFM than at KR.

my take

Yes, AMZN can expand the WFM customer base.  Yes, it can cross-sell, that is, deliver non-food goods, like Alexa, through the WFM store network and the 365 brand through the AMZN website.  Yes, using the Amazon store card will likely get customers a 5% rebate on purchases.  Yes, WFM’s physical stores may even serve as depots for processing AMZN returns.  That’s all gravy.

But if AMZN can eliminate what’s eating those 11.2 percentage points of margin (my bet is that it can do so in short order) it can lower food prices at WFM by a huge amount and still grow the chain’s near-term profits.  This is what I think activist investor Jana Partners saw when it took a stake in WFM.



A regular reader asked me the other day to explain why I said I thought the margins of Whole Foods (WFM) are too high.  Here goes:

what they are

Margins are ratios, usually some measure of profits (gross profit, operating profit, pre-tax profit…) divided by sales. (Yes, in cost accounting contribution margin is a plain old dollar amount, not a ratio.  But it’s an exception.  I have no idea why the misleading name.)

when high margins are bad

At first thought, it would seem that the higher the margins, the better off the seller is.  Buy the item for $1, sell it for $2.  That’s good.  Raise your prices and sell it for $5, that’s better.

The financial press encourages this notion with articles that talk up high margins as a good thing.

At some point, however, other people will work out how much you’re making and start doing the same thing.  They’ll typically go for market share by undercutting your prices.  So now you’ve got a competitor who wasn’t there before and you’re facing a price war that will at the very least undercut your brand image.

Creating what analysts call a price umbrella below which competitors can price their products and be protected from you as a rival is one of the worst mistakes a firm can make.

In my experience, it’s infinitely better to build a market more slowly by yourself than to have to try to dislodge a new rival who has spent time–and probably a lot of money–to enter.

One potential exception:  patented intellectual property (think:  Intel or drug companies). Even in this case, however, there’s the danger that once-successful firms become lazy and fail to continue to innovate after initial success.  The sad stories of IBM, or of Digital Equipment, or INTC for that matter, are cautionary tales.

More tomorrow.