The capital cycle is something every investor should be aware of

In common with many value investors, I spend most of my time analysing individual companies and very little time thinking about the economy or economic cycles.

I now realise that this is a mistake and that being aware of economic cycles, and the capital cycle in particular, could improve my investment returns.

The dangers of ignoring business cycles

A handful of my investments in recent years have been in the supermarket and commodity sectors and most of those investments have performed badly. An example would be Tesco, whose shares I recently sold.

It became obvious that there was a recurring pattern:

  1. I find a company with a good track record of growth and shares that are attractively priced
  2. The share price is attractive (i.e. low) because the company has recently run into what appear to be minor problems
  3. After a period of time, the minor problems become much worse and the company’s revenues, earnings, dividends and share price decline, sometimes dramatically
  4. It becomes clear that the company’s past growth will not be replicated in the future anytime soon and that the investment has become a value trap

Since I believe in the principle of continuous improvement I have spent a lot of time looking for some kind of signal or evidence that would have warned me that these investments had a high risk of becoming value traps.

Thanks to a bit of luck I think I have at last found something which could be the missing piece to this particular puzzle, and it’s called the capital cycle.

A brief introduction to the capital cycle

The economy is cyclical and that cycle is made up of other interacting cycles including the credit cycle, the sentiment cycle and the capital cycle, also known as the capital expenditure (capex) cycle.

I was reminded of the importance of the capital cycle by a presentation given by Edward Chancellor for the CFA UK earlier this year. The talk was based on his book Capital Returns, which is in turn based on a collection of letters from Marathon Asset Management to their clients.

As I watched the presentation online last Saturday it gradually dawned on me that the capital cycle could be just what I was looking for.

A short extract from the book should give you a good idea of what the capital cycle is all about:

“Typically, capital is attracted into high-return businesses and leaves when returns fall below the cost of capital. This process is not static, but cyclical – there is constant flux. The inflow of capital leads to new investment, which over time increases capacity in the sector and eventually pushes down returns. Conversely, when returns are low, capital exits and capacity is reduced; over time, then, profitability recovers. From the perspective of the wider economy, this cycle resembles Schumpeter’s process of “creative destruction” – as the function of the bust, which follows the boom, is to clear away the misallocation of capital that has occurred during the upswing.”

Or pictorially, from the same book:

Capital cycle
Most companies follow the capital cycle to some extent

So when a lot of capital is invested into an industry, either through increased capex or by the entrance of new competitors, the result is a lot of supply and a lot of competition, both of which tend to reduce returns on that invested capital.

But what does that have to do with my underperforming supermarket and commodity sector investments? I’ll try to explain using Tesco as an example.

Tesco is an example of the expansionary phase of the capital cycle

When I bought shares in Tesco in 2012 it had an impressive record of steady growth, which is exactly what I like to see:

Tesco results to 2012
Tesco had an impressive record of rapid and consistent growth

Back in 2012, I didn’t look at capex at all; it was just not on my radar at all. However, about two years ago I did start looking at capex and, more specifically, the ratio between a company’s capex and its post-tax profits.

This ratio is an important metric for me now because many capital-intensive companies:

  • Have higher fixed costs, which can make their profits and cash flows more volatile
  • Have to invest large amounts in new capital assets in order to grow, which can be a drain on cash and often involves taking on debt

This is how Tesco’s capex and profits looked in the run-up to 2012:

Tesco capex and profit 2012
Tesco consistently spent more on capex than it made in post-tax profit

Clearly, Tesco was investing more each year in capex than it made in post-tax profits. You can also see the impact of the financial crisis when capex was cut dramatically after 2009.

During that period Tesco:

  • Made £17.0bn in post-tax profits
  • Spent £28.9 billion on capital investments

This gave Tesco a ten-year capex ratio of 170% in 2012. I consider anything above 100% as “high” and Tesco was investing far more in capital assets relative to its profits than most other companies.

However, maintaining and opening supermarkets is a capital-intensive business and just having a high ratio of capex to profits doesn’t necessarily indicate whether Tesco was growing its capital assets or simply maintaining what it already had.

To understand that, we can look at the ratio between capex and depreciation.

The importance of the capex-to-depreciation ratio

In very simple terms, depreciation is the amount by which a capital asset loses value each year.

For example, imagine a company that buys a delivery truck for £10k and expects it to last 10 years, after which it will be worn out and worthless.

On the profit and loss statement expenses relate to the current financial year, so if the truck will be used for ten years it isn’t right to put the whole £10k down as an expense for the current year.

Instead, it would be better to account for the truck by recording the £10k as a capital expense on the balance sheet rather than a revenue expense on the P&L statement. The result would be a £10k capital asset rather than a £10k expense.

For the next ten years, the company would record a £1k depreciation expense, which would also reduce the value of the capital asset (i.e. the truck) by £1k until its value reached zero.

Since the value of the truck declines by £1k each year the company would do well to put aside £1k each year (and perhaps a little more) in order to buy a new truck when the old one wears out.

In some respects then, depreciation (and amortisation, which is the same thing but for intangible assets) can be seen as the ongoing replacement cost of a company’s capital assets.

If a company’s capital expenses are at about the same level as depreciation then the company is more or less replacing its existing capital assets, but not growing them.

For companies to grow they usually have to increase their capital assets and that means having capex significantly higher than depreciation, sometimes over many years.

In Tesco’s case, its capex and depreciation in the period up to 2012 look like this:

Tesco capex and depreciation 2012
More capex than depreciation means Tesco’s capital assets are expanding

Tesco’s capex in the run-up to 2012 was much higher than its depreciation rate. Having looked at the relationship between capex and depreciation for a few dozen companies, I can say that Tesco’s capital investment rate in that period is really quite exceptional.

Every single year it made capital investments far beyond the existing rate of depreciation; in other words, it was expanding its capital assets rapidly and investing aggressively for growth.

Tesco’s total capital expenditure over that period came to £28.9bn compared to total depreciation of £9.6bn.

In other words, Tesco invested an additional £19.3bn in capital assets and was investing at more than three times the rate required to simply maintain its existing assets (primarily supermarkets, fittings and fixtures, IT, supply chain assets etc.).

In my opinion, this is a clear indicator that Tesco was expanding its capital assets massively, bringing on huge amounts of new supply which would – according to the capital cycle theory – almost inevitably have a detrimental impact on profitability (especially when measured as return on capital employed).

What a capital cycle value trap looks like when the trap snaps shut

In 2012 some investors (including Neil Woodford) were exiting Tesco, perhaps because of the rise of the German discounters and competition in general.

But Tesco’s share price looked attractive and so I (along with Warren Buffett) bought some of its shares on the assumption that any slowdown would be a minor bump in the road and that the past record of growth would return in due course.

Events may have unfolded differently if I’d looked at Tesco from a capital cycle point of view. Having seen the enormous investment in capital assets I might have chosen to sit and wait for Tesco to enter the downward phase of the cycle.

If I had waited then I would have seen Tesco’s profits, dividends and share price collapse and would have avoided investing in this most high profile of value traps (although thanks to a policy of broad diversification the hit to my personal portfolio and the UKVI model portfolio was just a percentage point or two):

Tesco results to 2016
Tesco’s years of consistent success did not guarantee it a prosperous future

I think the massive capital asset expansion that Tesco and the other supermarkets embarked on during the previous decade(s) played a major role in the sector’s recent downfall.

Having reached the peak of the capital cycle, this is how Tesco looks now from the point of view of its post-tax profits and capex:

Tesco capex and profit 2016
Declining profits almost inevitably lead to declining investment in capital assets

As you can see, the collapse in profits has inevitably led to a massive decline in capex.

While I don’t have a crystal ball, this lower level of capex could easily last for many years given the huge expansion the company has gone through since the 1980s.

Looking at Tesco from yet another point of view, here’s how that lower level of capex compares to depreciation today:

Tesco capex and depreciation 2016
Capex is falling whilst depreciation is quite likely to keep going up

As the chart shows, unlike profits or capex (or dividends for that matter), depreciation does not decline so quickly or easily.

Once a capital asset has been added to the balance sheet it will typically depreciate at a fairly steady rate over many years, assuming big chunks of it aren’t sold off to pay down debts (which is something Tesco has also been doing recently).

One way to think of capital assets is that they’re like baby birds which constantly need feeding. As they get bigger they demand more food, and if you don’t feed them they’ll shrink and perhaps even die.

The problem for Tesco is that now it has built up this massive base of capital assets those assets need feeding with massive amounts of cash, and if they’re not fed they will fall into disrepair and generate even lower rates of return.

This process of capital consolidation may have begun as the chart shows Tesco’s 2016 capex falling below depreciation. This means its remaining capital base has started to shrink in value, finally ending a very long period of expansion.

With hindsight, it’s obvious that avoiding Tesco was the best option, given its massive and prolonged capital investment (there were other problems too, but here I’m focusing on the capital cycle). But hindsight is for historians. What I want to know as an investor is:

Was there some way of knowing in advance that Tesco was very likely to become a capital cycle value trap?

Using the capex-to-depreciation ratio to avoid capital cycle value traps

What I’m after is a ratio or other metric which is going to alert me to companies that are rapidly expanding their capital bases.

Having crunched the numbers for most of my holdings over the past five years and in particular those that have run into problems, I have decided to settle on the capex-to-depreciation ratio as my metric of choice:

  • Capex to depreciation ratio = capex / (depreciation + amortisation)

I’m interested in avoiding those companies where capex has been much higher than depreciation (and amortisation) over the last decade and so I’m looking to avoid companies where the capex to depreciation ratio is well above 100%.

But how high is “too high”?

There is no single correct answer, but from the research I’ve carried out, it seems that those companies where the capex to depreciation ratio is consistently above 200% are the ones that are much more likely to run into problems later on, caused in part by excess investment and excess supply.

So taking account of both capex in individual years and over the last decade as a whole, this is my new rule of thumb for avoiding capex cycle value traps.

New rule of thumb:

  • Only invest in a company if its capex-to-depreciation ratio for the last ten years as a whole is below 200% and if the ratio is below 200% in more individual years than not during that period

Applying that rule of thumb to Tesco:

  • Tesco’s overall capex-to-deprecation ratio for the period leading up to 2012 was 299%
  • Tesco’s capex to depreciation ratio was over 200% in ten years out of ten in the run-up to 2012

Clearly, Tesco was flashing all sorts of warning signs in terms of its potential to be a capital cycle value trap.

There were many other warning signs too, which I’ve already covered in my post-sale review, but the company’s massive investment in capital assets is an important one.

Other examples of my investments which (with hindsight) had a high risk of becoming a capital cycle value trap were:

  • BHP Billiton (purchased in 2011): In 2011 BHP had a ten-year capex-to-depreciation ratio of 241% and the ratio was above 200% in seven of those years
  • Rio Tinto (purchased 2012): In 2012 Rio Tinto had a ten-year capex-to-depreciation ratio of 252% and the ratio was above 200% in seven of those years
  • Wm Morisson (purchased 2013): In 2013 Morrison had a ten-year capex to depreciation rate of 207% and the ratio was above 200% in five of those years, so Morisson was a borderline case in terms of the capital cycle

I still hold each of those companies and so whether they will eventually turn out to be good or bad investments is currently unknown. However, so far they have each performed terribly.

Having bought the companies a few years ago, each has since moved from the expansionary phase of the capital cycle through to the consolidation phase and that has occurred alongside collapsing profits, dividends and share prices.

At the very least, waiting for the capital cycle to turn would have resulted in me either not investing at all or eventually investing at a significantly lower share price than the one I actually paid.

However, it’s not all doom and gloom.

The vast bulk of my current and previous investments easily make it past that capex/depreciation rule of thumb and have not had any noticeable capital cycle-related difficulties to date.

On that basis, I think this new rule of thumb is a reasonably good first stab at producing a capital cycle value trap warning signal.

Of course, if it turns out to be too restrictive, or too lax, I will adjust accordingly, but for now, I’m happy to add this ratio to my existing investment toolbox.

If you want to know a whole lot more about the capital cycle I suggest you buy the Capital Returns book or at the very least download this sample chapter (PDF).

Finally, there’s a video interview at MoneyWeek between Merryn Somerset Webb and Edward Chancellor covering the capital cycle here.

Author: John Kingham

I cover both the theory and practice of investing in high-quality UK dividend stocks for long-term income and growth.

19 thoughts on “The capital cycle is something every investor should be aware of”

  1. Wouldn’t it be better to apply this ratio to a whole sector instead of a single company. There could be cases where some competitors shrink while others expand leaving the overall capacity static.

    1. Nukular,

      Indeed it would. In fact the Capital Returns book makes that exact point. However, there are a few reasons why I chose to not formally apply the ratio to sectors as well as individual companies:

      1) From the research I have carried out there does appear to be a good correlation between companies that grow their capital assets quickly and those that have subsequent excess supply problems. Looking at the sector level does not appear to make any difference in these cases
      2) I am lazy (or efficient) and analysing the data for a whole sector could be a lot of work
      3) Capital can flow into and out of a sector through new entrants or competitors exiting, so it’s very difficult to measure capital in/outflow for a whole sector

      However, your point is still valid.

      What I will probably do in practice is have a quick look at a few competitors of any company I’m analysing to see if there is any hint of sector-wide capital expansion. Also perhaps looking at news about the sector overall (which I do anyway) to help me think about which stage of the capital cycle it might be in.

  2. You may be on to something here with this new measure.

    Terry Smith you may recall avoided Tesco based on declining ROCE.
    How the two inter-related measures match up is not entirely clear?
    But then two measures can be more revealing than one.
    Measure for Measure!

    P.S. Don’t beat yourself up about Tesco, we too were firmly in that ‘hold’ camp for many years. And Warren Buffett buying in just confirmed that conviction!

    1. Hi Magneto, the relationship between capex/depreciation and ROCE is that ROCE is high, which makes large capital investments seem sensible (which it may be, to some extent). Eventually supply exceeds demand and ROCE starts to drop. Revenues, profits and capex decline, until the existing capital assets are worn out, at which point demand exceeds supply and so ROCE starts to go back up again. That’s a very simplified and idealised picture of reality of course.

      As for Tesco, I’m actually happy to have invested in it as it taught me a great deal, and those lessons will earn me far more than I lost on that one investment.

      So always look on the bright side of life… (as somebody once said).

  3. Hi John, Referring to The Restaurant Group, should we also consider companies with large operating leases, which are in fact capital assets (such as renting premises), and apply some kind of calculation to these leases. I know it will depend on what their obligations are contractually,but is it something that perhaps should be reviewed?

    1. Hi Geoffrey, lease obligations are an interesting one. Some investors do like to ‘capitalise’ lease obligations, effectively turning them into something like a mortgaged property, with a capital asset purchased by debt. It’s a reasonable thing to do and the main person I know who does this is Richard Beddard at Interactive Investor.

      Here’s Richard’s write-up of Games Workshop.

      However, I don’t currently capitalise lease obligations. So far none of my investments have run into problems caused by their leases, even though I have owned a few retailers (primarily JD Sport, which worked out quite well).

      In my limited experience with lease-heavy companies, leases are flexible enough that they don’t cause major problems, even if a company is struggling for a while. So Greggs or JD Sport, when they went through tough periods, were able to let unprofitable stores close when their leases ran out, or even renegotiate the lease with the landlord, probably for a one-off fee. Bank debts, in my experience, have the capacity to be a more ‘systemic’ risk, i.e. at the level of the company, rather than at the level of an individual store.

      So until I make an investment which then has problems which are caused directly by using a leased property business model, I’m going to hold off with capitalising leases.

      Having said that, investors should at least be aware when a company they’re investing in has a lease-heavy business model.

  4. John – interesting article my only thoughts would be that this works where the organization is reasonably capital intensive but many are not. Their ROCE would be much higher but the amount invested would be relatively low as would depreciation. So I do believe their is an industry bias here that would be important?

    1. Hi Nick, yes you’re right, thanks for pointing that out.

      This does apply more to capital intensive industries. So in practice this new capex/depreciation rule of thumb will be used in combination with the existing capex/profit rule of thumb, where I’m a bit more cautious about companies that have spent more on capex than they make in profits.

      For example, Admiral Insurance (which as you know is in the UKVI model portfolio) has a ten-year capex/depreciation ratio of more than 200%, but on average its capex is only about 8% of its profits, so it’s a very capital-light business, which is what you’d expect of an insurance company (all they need is a call centre – a few desks and telephones – in Wales, more or less).

      Admiral only has three out of ten years where capex/depreciation is above 200%, so it doesn’t quite fail the capex/depreciation rule of thumb, but it’s a close run thing. However, I wouldn’t be worried in practice because Admiral’s overall capex is so low relative to profits. So although it is investing and expanding its capital assets quite quickly, the growth of those assets does mean it is massively increasing the supply of car insurance.

      So yes, there is an industry bias and for insurance, as one example, the capital cycle does not apply in the same was as for companies that need a lot of physical assets to supply their products or services. I would expect the same sort of thing applies to banks, asset management, software, support services and other sectors where physical infrastructure is not a major expense.

  5. Where do you get the capex, depreciation and amortisation data for 10 years from – do you have to search back through 10 years’ worth of annual reports?

    1. Hi Tim, I get the figures from SharePad. It allows you to download more than ten years of income, balance sheet and cash flow data as a spreadsheet, after which it’s easy to do these various calculations.

      You can sign up to SharePad for a 30-day trial and get one month free using this link (it’s a ‘refer a friend’ link, so I’ll get a free month as well):

      I’m going to add this capex/depreciation ratio to the spreadsheets on the ‘free resources’ page of this website soon, although having said that the calculations aren’t particularly difficult.

    2. I think other data providers such as Morningstar (their Premium service) might have this too. Otherwise yes, it’s a case of trawling through the annual results.

      1. Thanks, I’ve been thinking of having a look at ShareScope/Sharepad for data for my UK shares. I’ll use your link if I decide to give them a try.

  6. John, I guess avoiding very “Capital Intensive” companies altogether is a first port of call.

    But if you must, isn’t identifying this really synonymous with watching the build up of debt and the relative decline in profit / operating margins and cash flow. Both of these things were present in Tesco’s case. I think watching these is probably as significant, as is the realisation that the food delivery market was drowning in overcapacity. The thing that did it for Tesco, was watching all the supermarkets being built — in the space of a few years each town went from 1 to 3 to about 5 to 10. That’s probably a layman’s understanding of the capital cycle.

    One example of declining operating profits and operating margin’s is Capita — which has accrued a huge debt relative to steady declining profits, and it seemed pretty obvious that it’s share price would fall and significantly. I reviewed this 2 years ago and one year ago and came to the conclusion it was staring one in the face.
    This argument probably doesn’t fit into the direct context of measuring the upward or downward trend in capital expenditure or the relative size of capex spend compared to the companies business. It simplifies what you are looking at though :-

    Capita profits 5 years ago = £355M — today £206M, despite growing revenue 65% in that period. The company is running forward at pace to travel backwards.
    During that period debt (borrowings) rose from £1,695M to £2,163M a rise of 28%
    The ratio of profits to borrowings is off the scale — 2163/206 == 10.5X and 39X if you take after tax at £55M.
    Operating margins have dropped from 12 to 4% in that 5 year period.

    The share price has already fallen from 1350 to 976 in the last 12 months.

    This is one of Neil Woodford’s top 10 holdings, why?
    I note that in May BT was sold out of the fund.


    1. Hi LR, yes generally I try to avoid high capex companies, or if not avoid then at least insist on other aspects of the business being lower risk, such as debt levels.

      As for the synonymity (my spell checker assures that’s a real word) of increasing debts and decreasing profits/margins/cash flows, I would say yes, to some extent. Not necessarily increasing debts, but the latter stages of the capital cycle tend to show up as decreasing return on equity, as increased supply in the sector starts to exceed demand.

      So you’re right, and looking at more than one metric is always a good idea; so in combination with the capex/depreciation ratio I would also be looking at the evolution of ROCE over a few years, along with debt and also supply across the sector.

      Looking at Capita on my stock screen quickly confirms what you’ve said: its debt ratio (borrowings/five-year avg profits) is almost 10, whereas my limit for Support Services companies is 4… so it has way more debt than I would be happy with. On that basis alone I wouldn’t touch it with the proverbial barge pole.

      I’m sure Mr Woodford has his reasons, but they must be very different to yours or mine.

  7. HI John
    extending this line of thought a little it could also indicate a prudent point to terminate a particular investment and move on. Do you intend to make any use of this when assessing the next disposal?

    1. Hi Stephen, yes you’re right, as well as being relevant to any purchase decision the capital cycle is also relevant when deciding what to sell, so I’ll be factoring it in, much as I do with the other ratios I use.

      In most cases I would expect it to have a marginal impact on what I decide to sell, but occasionally it might be lead to me selling something that I might ordinarily want to hold on to.

      For example, a mining company which is expanding quickly, with high profitability, but which is also investing massively in capital assets (along with others in its industry) in order to expand supply.

      The high growth rate and profitability would normally make me want to hold, unless the valuation was sky high. But a very high rate of capex and capital asset expansion might now make me more wary of holding on to what could turn out to be a ticking time bomb. It would be better to exit during the expansion phase of the capital cycle rather than wait (unintentionally) for the cycle to turn downward.

  8. Would you consider R&D to be a part of the capex bill in this context?

    1. Hi Andy, no, not in this context. The capital cycle is mostly about investment on the supply side which is sticky, i.e. new mines, oil wells, factories, stores and other capital assets which then have an a relatively fixed overhead cost and supply capacity which cannot easily be switched off when demand fails to keep up with expanding supply.

      R&D could lead to the development of sticky supply, but the link isn’t nearly as direct as it is with spending on property, plant, equipment and so on.

Comments are closed.