Why dividend investors should look at free cash flow

If you’re a dividend investor, I think you should pay more attention to a company’s free cash flow than its earnings.

In a nutshell, here’s why:

Dividends are paid out of cash, so we need to make sure a company is consistently generating more than enough free cash (i.e. spare cash) to pay the dividend.

And earnings (and therefore the standard dividend cover ratio) aren’t always a good indicator of free cash generation.

As a quick reminder, here’s my preferred definition of free cash (there are other definitions, but this is the one I use):

Free cash flow = Net cash from operations (net of interest and tax) - capital expenses

What’s so bad about earnings, I hear you ask?

For the most part, not much. But earnings can sometimes be misleading, especially when they deduct significant non-cash expenses and ignore significant cash expenses.

Deducting significant non-cash expenses can lead to earnings that are much smaller than the actual amount of free cash generated by the company. On the other hand, ignoring significant cash expenses can leave earnings much higher than the amount of free cash generated.

In many cases, these differences don’t really matter and earnings and free cash generation are quite similar.

But for some companies, there is a large gap between the earnings they report and the free cash they generate.

In those cases, focusing on earnings and an earnings-based dividend cover ratio can be misleading at best and disastrous at worst.

To explain this in a bit more detail I’ll use a couple of examples, the first of which will focus on a company whose free cash generation is much higher than its reported earnings.

Imperial Brands has free cash flows which are significantly higher than earnings

Over the past decade, Imperial Brands (formerly Imperial Tobacco) generated average earnings per share of 123p.

Compare that to the company’s average dividend of 126p per share and you can see that earnings have just about failed to cover the dividend over ten years.

Looking at the most recent five-year period it’s even worse, with an average EPS of 136p and an average dividend of 157p.

And the 2018 result was worse still, with reported earnings of 146.3p and a dividend of 187.8p.

From an earnings perspective, it seems as if the company is skating on thin ice. Over a ten-year period the company’s earnings have failed to cover the dividend and the situation is gradually becoming worse.

If we look at a chart of Imperial Brands’ earnings and dividends we can see more clearly what’s going on:

Over the last decade, the company’s dividend has consistently increased by around 10% each year, and yet its earnings have gone nowhere.

Most dividend investors would look at this chart and conclude that the company was a high-risk yield trap (Imperial’s dividend yield is currently well into the ‘red flag’ zone at more than 8%).

However, from a free cash flow point of view, everything appears to be fine. In fact, over the last ten years, Imperial Brands’ free cash flows have more than covered the dividend in every single year. Just look at the chart below:

In this second chart, we can see that free cash flows have been consistently higher than the dividend. This suggests that the company is generating more than enough cash to afford those increasing dividend payments.

So why is there such a big difference between earnings and free cash flows? The answer has to do with acquisitions which, for the most part, were made more than ten years ago.

How Imperial Brands earnings are reduced by the amortisation of acquired goodwill

After re-joining the stock market as an independent company in 1996, Imperial Brands (or Imperial Tobacco as it was then) set out on an ambitious strategy to grow through a series of very large acquisitions.

Over 12 years to 2008, the company spent upwards of £17 billion acquiring other companies, eventually becoming the fourth largest tobacco company in the world.

As is often the case with acquisitions, Imperial paid more for the companies it acquired than the book value of its tangible assets.

In other words, a company might have £100m of tangible assets on its balance sheet, but if it’s highly profitable and earns £20m each year its value to a buyer would likely be more than £100m.

If a buyer wanted to get a 10% return on their acquisition then they might pay £200m for that company (i.e. the company’s £20m annual profits would give the buyer a 10% return if the acquisition price was £200m).

If that acquisition went ahead, the buyer would add the acquired company’s £100m of tangible assets to its balance sheet. The £100m paid above and beyond the cost of those tangible assets would be recorded on the acquirer’s balance sheet as an intangible ‘goodwill’ asset.

If the acquisition was made prior to 2005 (as most of Imperial Brands’ were) then that goodwill asset would be amortised (effectively depreciation for intangible assets) over several decades.

This amortisation expense is of course a non-cash expense because most of these companies were acquired more than ten years ago and there is no actual cash outlay.

So the amortisation of acquired goodwill has no effect on free cash flow, but it does reduce earnings.

Most of the time the effect is quite small, but in the case of highly acquisitive companies, it can make a huge difference.

For example, Imperial Brands’ earnings are reduced by about £1 billion each year by this non-cash expense.

£1 billion sounds like a lot, and it is. The company’s reported profits are currently around £1.5 billion, and without that £1 billion amortisation expense profits would be almost 70% higher at £2.5 billion.

That’s a massive difference and it’s the main reason why Imperial’s free cash flows are so much higher than its reported earnings.

My second example will focus on a company which has the opposite problem, i.e. its free cash generation is much lower than its reported earnings.

Ted Baker has free cash flows which are significantly lower than earnings

Unlike Imperial Brands, Ted Baker generates reported earnings far in excess of its rapidly growing dividend (here’s a detail review of Ted Baker, and the company is currently a holding in my personal portfolio and model portfolio).

The chart below says it all:

However, along with earnings and dividends, something else which is rising rapidly at Ted Baker is its debt.

Until recently, the company had never borrowed more than the odd million here or there. Since 2011 that’s all changed, and the company’s debts have gone from zero to almost £130m in just a handful of years.

Okay, for a company producing net profits of about £50m, borrowings of £130m are in no way excessive, but if they keep going up at their current rate that won’t be true for much longer.

So why does a company which generates earnings far in excess of its dividend need to ramp up borrowings so aggressively and uncharacteristically?

The answer can be found by looking at the company’s free cash flows, which are starkly different to its earnings:

On the chart above, Ted Baker’s free cash flows look like something out of a Loch Ness Monster movie; rising above the waterline only to disappear shortly after into the murky depths below.

Obviously something is going on here. With Imperial Brands we had a company where earnings were consistently lower than free cash flows because of the large, recurring non-cash expense of goodwill amortisation.

Ted Baker has the opposite problem; a large, recurring cash expense which reduces free cash flows but doesn’t reduce reported earnings.

How Ted Baker’s free cash flows are reduced by capital expenses

Ted Baker is a rapidly growing company. According to my Growth Rate metric, it’s grown by almost 20% per year for the last ten years.

And in order to grow, companies have to invest in their tangible assets (yes, some companies grow by investing in intangible assets such as software and patents, but they still have to invest in tangible assets as well).

As you might expect, these asset’s aren’t free. They need to be funded by cold hard cash and how these costs are accounted for is where it all starts to get a bit murky.

An example might make things a little clearer, and in my article on the Growth Rate metric I wrote about a small hypothetical barber’s shop which we started together.

For the barber’s shop to function, we needed some tangible assets, including a barber’s chair.

Let’s say the barber’s chair cost £10k (it’s a nice chair). We paid for the chair in cash, so it has an immediate negative impact of minus £10k on our cash flows, but from an earnings point of view it’s not so simple.

That’s because listed companies use accrual accounting, where income is reported only when it’s earned and expenses are reported when the related service or product is used (that’s the basic picture, although there are a million caveats which I have neither the time nor inclination to go into…)

For example, at year end our barber shop may not have paid its electricity bill as these are paid in arrears rather than in advance. However, we have used the underlying product (electricity) so we still have to record the electricity expense even though we haven’t paid for it in cash.

The opposite happens when we buy the barber’s chair. We pay out £10k in cash today, but if the useful lifetime of the chair is ten years, we haven’t ‘used’ that asset, so we don’t record the expense on the income statement. Instead we enter it as an asset on the balance sheet.

This leaves earnings unaffected by the £10k capital expense.

Obviously, barber’s chairs don’t last forever and we have to record it as an expense at some point, so companies typically record a non-cash depreciation expense each year using what’s known as the ‘straight line method’.

In our case that means depreciating the barber’s chair by £1k each year for ten years, and reducing the asset’s value on the balance sheet by the same amount.

This non-cash depreciation expense reduces earnings but doesn’t reduce cash flow because it’s a non-cash expense.

So we have two sides of the same coin:

• Capital expenses (capex) – which reduce cash flows but not earnings, and
• Depreciation – which reduces earnings but not cash flows.

If annual capex equalled depreciation then there would be no problem because the two would cancel each other out. Capex would reduce cash flow and boost earnings and deprecation would boost cash flow and reduce earnings by the same amount.

Earnings would more or less equal free cash flows and life would be easy.

But in reality, capex is typically higher than earnings. There are two reasons:

1) Inflation drives up the cost of replacing capital assets

In our barber shop example, let’s say we whizz forwards ten years into the future and we need to buy a new chair. Even with inflation at just 2% per year, a new chair is likely to cost at least 20% more than an equivalent chair did a decade ago.

This means our current capital expense (to buy the new chair) is £12k whereas our depreciation over the last decade (i.e. the cost of the old chair) came to £10k.

So as long as there’s inflation, capital expenses to buy tomorrow’s assets will usually be higher than the depreciation cost of yesterday’s assets, and this is one reason why average free cash flows are typically lower than average earnings.

2) Growing companies need to grow their fixed assets

Now imagine that our barber’s shop is doing well and growing. At some point we max out the number of people we can get into a single barber’s chair each day. Our barber works seven days a week, 18 hours per day and still we have a long queue of customers out the door.

Clearly, our business’s primary bottleneck is the fact that we have one barber chair. The obvious solution is to buy a second chair, and that simple fact illustrates why the gap between capex and depreciation is even larger for growing companies.

Remember, our depreciation expense over the last decade came to £10k as we reduced the balance sheet value of our barber’s chair to zero.

But now we need to buy two chairs. One to replace the worn-out original and one to allow the business to grow.

Each chair costs £12k (up from £10k thanks to inflation) so we need to spend £24k as a capital expense, not the £10k we spent a decade ago.

If our business was growing even faster we might want to buy three chairs, which would cost £36k. As you can see, the faster a company is growing the more cash it has to spend on capital expenses today, and the greater the gap becomes between the capital expense of future assets and the deprecation expense of historic assets.

Let’s forget about the barber’s shop for a minute and focus on Ted Baker.

Ted Baker reinvests virtually all its net operating cash back into the company to drive future growth

Ted Baker is a reasonably profitable company and generates cash from operations equal to about 18% of the value of its tangible assets.

If our barber shop produced an 18% annual return our investment in its assets I’d be pretty happy.

But not all of that cash generated makes it back into the hands of shareholders. In fact, almost none of it does.

For example, over the last decade Ted Baker generated net cash from operations (net of tax and interest payments) of £280m.

Of that £280m, £275m was reinvested back into the business to fund growth. This includes the purchase of its London headquarters, the ‘ugly brown building’ in King’s Cross, but most of Ted Baker’s capex reflects the relatively high cost of rolling out more and more of the company’s notably ‘eccentric’ (and lavishly decorated) stores.

You can see the purchase of the company’s HQ in 2016 as a spike in the chart below. But even outside that exceptional year, capex has eaten up pretty much every penny of cash from operations for at least the last decade:

£280m of net cash from operations over a decade, minus £275m of capital expenses leaves just £5m over an entire decade to pay the dividend.

However, Ted Baker actually paid out dividends totalling £132m over those ten years, an amount which looks eeirely similar to the £130m debt pile it built up over the same period.

From this free cash flow point of view, it begins to look like Ted Baker is funding its dividend almost entirely with debt.

In the short-term that isn’t a problem, but as a long-term strategy it’s a complete non-starter.

Having said that, I don’t think this makes Ted Baker a bad company. As far as I can tell it’s a solidly profitable company which is (perhaps quite sensibly) reinvesting heavily to drive future growth.

What I don’t think is a good idea is a rapid increase in the dividend, to the extent that a sizeable chunk of debt has had to be taken on in order to afford it.

Personally I still like Ted Baker, but I think management should:

1. Reduce growth capex in order to free up some cash to cover the dividend, and
2. Reduce the dividend’s growth rate below that of operating cash flow so that cash generation has a chance to catch up with and overtake the dividend.

Free cash flows give investors a better insight to dividend sustainability

In summary then, from an earnings point of view, Imperial Brands appears to have an unsustainable dividend which is uncovered by earnings, while Ted Baker appears to have a robust and sustainable dividend covered more than twice over by earnings.

In both cases I think this earnings-based viewpoint is misleading.

By looking at free cash flows we get a different picture; one where Imperial Brands has a sustainable dividend more than covered by free cash flows, whereas Ted Baker appears to have an at-risk dividend which hasn’t been covered by free cash flows for years.

I think a free cash flow-based analysis of dividend sustainability is likely to be more accurate than an earnings-based analysis, and that’s why free cash flows will be replacing earnings in all of my metrics from 2019 onwards.

Author: John Kingham

I cover both the theory and practice of investing in high-quality UK dividend stocks for long-term income and growth. View all posts by John Kingham