### Defensive Value Investing: A 20-week course

**2: Growth**

“The stock of a growing company, if purchasable at a suitable price, is obviously preferable to others.”

Benjamin Graham

Last week we looked at how to find companies with consistent histories of profits and dividends. That’s a reasonable starting point, but there are many more things that we need to look for.

Thanks to inflation, unless a company can grow its earnings and dividends over time then in real terms the value of its economic output will fall. This means that in addition to consistent profits and dividend payments, I always look for growth.

Growth can be broken down into two components – quality and speed – and this week we’ll look at both of these in turn.

## Measuring growth quality

One way to measure the quality of a company’s growth is to measure the consistency of its growth. In other words, how often the company grew over, say, the last ten years.

Measuring growth quality in terms of consistency makes a lot of sense because, in general, companies with reasonably consistent growth in the past are more likely to produce growth in the future. And since much of our investment returns will come from share price growth as well as dividend income, I think it’s a good idea to invest in companies where future growth is more likely.

*But companies produce hundreds of different financial figures. Exactly what should we measure the growth quality of? *

I think we should measure growth quality across three of a company’s most fundamental inputs and outputs: Revenues, earnings and dividends. Let’s start with dividends.

For most established companies, dividends are the primary source of returns from both income and capital growth over the long-term (e.g. ten-years or more). Given their central importance to returns, it makes sense to measure the consistency (or lack thereof) of their growth. And if you’re not sure what dividends have to do with capital growth, just remember this:

As a company’s dividend grows, it’s dividend yield will grow too. That will eventually attract more investors and this increase in demand will push the share price up. This will go on until the dividend yield is no longer attractive enough to pull in additional investors. This is obviously oversimplified, but it’s a fair representation of how dividend growth drives capital growth.

So dividends are important, but as we saw in Chapter 1, dividends are paid out of retained earnings, so if dividends are to grow over the long-term, so must earnings. And rather than volatile and inconsistent earnings growth, I would rather invest in a company with steady, consistent and (relatively) predictable earnings growth.

Last but not least, we have revenues. Measuring revenue growth quality is important because earnings are what’s left over from revenues after all expenses are deducted, so if earnings are to grow consistently over the long-term, then so must revenues.

In summary then, I look for progressive and consistent growth across revenues, earnings and dividends, and I combine these into a single metric which I call the **growth quality** score.

### Calculating growth quality

To calculate the growth quality score we need a company’s revenues per share, earnings per share and dividends per share going back over the last ten years. EPS and DPS are usually quoted in a company’s annual results, but revenues are usually stated as a total figure (e.g. £100 million) so you’ll probably have to calculate revenue per share (RPS) as:

revenue per share = revenue * 100 / number of shares

You can find revenues and the number of shares in each company’s annual results (I use Diluted Weighted Average Shares which you can find in the accounting notes at the back of each annual report).

Once we have the data, we can take the following steps to calculate growth quality:

Steps to calculate growth quality

1) Count how many times revenue per share went up in the last ten years

2) Count how many times earnings per share went up in the last ten years

3) Count how many times dividends per share went up in the last ten years

4) Count how many times the dividend was covered (EPS greater than DPS)

5) Add up the results from 1, 2, 3 and 4 above

6) Convert to a percentage

Don’t worry if that looks complicated as we’ll be working through an example shortly. And also don’t forget there are links to spreadsheets which can do the calculations for you on the Free Resources page.

Once again I have a rule of thumb for growth quality, which should help to rule out companies whose growth is too volatile for a relatively defensive portfolio.

Rule of thumb

Only invest in a company if its growth quality is above 75%.

*Let’s have a look at how the growth quality score works in the real world with an example.*

### Burberry’s growth quality

I’ll use Burberry because it’s a good example of a company with high growth quality. Table 2.1 shows Burberry’s revenue, earnings and dividends per share up to its 2018 annual results.

Year | Revenue PS (p) | Earnings PS (p) | Dividend PS (p) |
---|---|---|---|

2009 | 271.3 | – 1.4 | 12.0 |

2010 | 274.8 | 34.4 | 14.0 |

2011 | 339.7 | 48.3 | 20.0 |

2012 | 418.3 | 59.3 | 25.0 |

2013 | 449.9 | 57.0 | 29.0 |

2014 | 521.8 | 72.1 | 32.0 |

2015 | 564.1 | 75.1 | 25.2 |

2016 | 561.6 | 69.4 | 37.0 |

2017 | 620.0 | 64.9 | 38.9 |

2018 | 636.4 | 68.4 | 41.3 |

**Table 2.1: Burberry’s ten-year results to 2018**

I don’t think it’s very easy to get a feel for the quality of Burberry’s growth just by looking at a table of numbers like this. Everything seems to be going up steadily, but earnings were negative in 2009, so there are definitely some issues lurking in the data. This confusion is precisely the sort of problem the growth quality score is designed to solve, or at least reduce.

By converting all those revenue, earnings and dividend figures into a single growth quality number it becomes much easier to understand how consistent Burberry has been; that in turn makes it much easier to compare Burberry to other companies.

I’ll calculate Burberry’s growth quality score step by step to show you how it works:

Calculating Burberry’s Growth Quality

1) Count how many times revenue per share went up in the last ten years

*revenue increase = 8 times*

2) Count how many times earnings per share went up in the last ten years

*earnings increase = 6 times*

3) Count how many times dividends per share went up in the last ten years

*dividend increase = 9 times*

4) Count how often the dividend was covered (EPS greater than DPS)

*dividend covered = 9 times*

5) Add up the results of 1, 2, 3 and 4 above

*8 + 6 + 9 + 9 = 32*

6) Convert to a percentage

*growth quality = 32 / 37 × 100% = 86.5% *

86.5% is a very high growth quality score. It reflects the fact that Burberry has produced some of the most consistent and broad-based growth of any FTSE 100 company.

Also, note that growth quality as a number is divided by 37 to turn it into a percentage. 37 is the maximum possible score (nine increases each of revenues, earnings and dividends per share, plus dividends covered ten times), so a company with a growth quality number of 37 would have a growth quality score of 100%.

## Measuring the long-term growth rate

Benjamin Graham said that a growing company is preferable to others, which seems obvious enough. I would go one step further and say that a high-growth company is preferable to a low-growth company (all else being equal, of course).

In order to differentiate between the hares and the tortoises I use a measure of growth which is:

**Broadly based**, reflecting the company’s overall growth as much as possible**Long-term**, measuring growth over a decade rather than growth over a year or two

### Long-term revenue, capital employed and dividend growth

As with growth quality, I measure a company’s growth rate by combining the growth rate of its revenues and dividends over a ten-year period. The idea is to use the past as a guide to the company’s sustainable future growth rate, which will always be impossible to know exactly.

One difference between my growth quality and growth rate metrics is that I measure capital employed growth instead of earnings growth. Before I explain why, here’s a definition of capital employed and its component parts:

Definition

**Capital employed: **The sum of equity capital (shareholder equity) and debt capital (total borrowings). I’ll abbreviate Capital employed per share as CEPS.

**Shareholder equity:** Equal to the shareholders’ claim on company assets. Can be calculated as total assets minus all non-shareholder liabilities. It is also equal to funds raised from shareholders plus retained earnings minus payments to shareholders (e.g. dividends). Can be found on the balance sheet.

**Borrowings:** The sum of current and non-current interest-bearing debts owed to banks and other lenders. Can be found on the balance sheet.

The problem with earnings is that they can be very volatile and occasionally negative. This can make measuring the earnings growth rate very hard because unusually high or low earnings at the start or end of the period being measured can have a major impact on the calculated growth rate.

To get around this problem, Benjamin Graham proposed measuring growth between the average of the oldest three years in a ten-year period and the average of the latest three years. By doing that, the earnings in any one year become less important and the general change over a number of years becomes more important.

This works to a degree, but after several years of using Graham’s technique to smooth out earnings growth, I still felt earnings were too volatile to be useful.

What I wanted was a factor that is fundamental to a company’s progress and yet, like revenues and dividends, doesn’t change dramatically from one year to the next, at least most of the time.

The factor that best fit the bill was capital employed.

Capital employed is money which has effectively been raised from shareholders and debt holders and is typically invested into productive assets such as property, factories (often called plant) and equipment. And if a company wants to double its revenues, earnings and dividends then, in most cases, it will have to double its productive assets and therefore its capital employed as well.

So if revenues are the source of all earnings and dividends, capital employed is the foundation upon which revenues are built.

To make life simple I’ll refer to this somewhat complicated growth rate metric as a company’s **growth rate**.

### Calculating a company’s growth rate

To calculate a company’s growth rate we’ll need its revenue and dividends per share for the last ten years, which you should already have if you worked through the Profitable Dividend calculations from last week.

You’ll also need capital employed per share for each of the last ten years, which can be calculated as below (assuming equity, borrowings and number of shares are quoted in millions in the annual results, which is usually the case):

capital employed per share = (total equity + total borrowings) * 100 / average number of shares

The steps for calculating a company’s growth rate are:

Steps to calculate Growth Rate

**1) Calculate revenue per share growth**:

1.1) Calculate average revenues per share for the oldest three-year period

1.2) Calculate average revenues per share for the latest three-year period

1.3) Calculate revenue per share growth:

*revenue growth = (new average RPS / old average RPS) – 1 × 100%*

**2) Repeat step 1 for capital employed per share**

**3) Repeat step 1 for dividends per share**

**4) Calculate total growth as the average of RPS, CEPS and DPS growth:**

*total growth = (RPS growth + CEPS growth + DPS growth) / 3*

**5) Calculate the growth rate required to produce the total growth (expressed as a spreadsheet function):**

*growth rate = (POWER(100% + total growth), 1/ 7) – 100%)*

If you’re wondering what the 1/7th part of the growth rate calculation is doing, it’s raising the total growth figure to the 1/7th power, which is the opposite of raising it to the seventh power.

*Why do we do that?*

Imagine if we started with the annual growth rate and wanted to know what total growth that growth rate would produce over seven years. To do that we’d 100% to the annual growth rate and raise the result to the seventh power (i.e. times it by itself seven times).

However, we’re actually starting with the total growth over seven years (there are seven years between the old and new three-year average periods) and we want to know the annual growth rate required to produce that total growth. In other words, we’re doing the opposite calculation, and the opposite of raising a number to the 7th power is to raise it to the 1/7th power.

This sounds more complicated than it actually is, and a spreadsheet can do this sort of calculation in an instant. But rather than get bogged down in theory, I think an example is the best way to show how this all works. But first, my growth rate rule of thumb:

Rule of thumb

Only invest in a company if its growth rate is above inflation (approximately 2%).

*Let’s see how this one works in the real world.*

### Burberry’s growth rate

I’ll use Burberry as an example one last time because it will allow you can see the difference between the company’s earnings (which, as you can see in Table 2.1, were negative in 2009) and its capital employed. In this case capital employed is, as I’d expect, more stable than earnings, even for a relatively steady growth company like Burberry.

Year | Revenue PS (p) | Capital employed PS (p) | Dividend PS (p) |
---|---|---|---|

2009 | 271.3 | 177.1 | 12.0 |

2010 | 274.8 | 184.7 | 14.0 |

2011 | 339.7 | 199.6 | 20.0 |

2012 | 418.3 | 242.3 | 25.0 |

2013 | 449.9 | 258.1 | 29.0 |

2014 | 521.8 | 293.0 | 32.0 |

2015 | 564.1 | 327.8 | 25.2 |

2016 | 561.6 | 361.0 | 37.0 |

2017 | 620.0 | 387.1 | 38.9 |

2018 | 636.4 | 336.2 | 41.3 |

**Table 2.2: Burberry’s results for the ten years to 2018**

You can get a sense of how fast Burberry has grown by looking at its 2018 revenues, capital employed and dividends per share compared with the values for 2009. All of them have more than doubled in that time (although capital employed took a step backwards in 2018), so the company has definitely grown quite quickly.

However, as with growth quality, it will be much easier to understand Burberry’s growth rate and compare it to other companies if the mass of figures in Table 2.2 is first converted into a single number.

Following the steps outlined previously, here is the calculation for Burberry’s growth rate:

Calculating Burberry’s Growth Rate

**1) Calculate revenue per share growth:**

1.1) Calculate average RPS for the old three-year period:

RPS for 2009, 2010 and 2011 was 271.3, 274.8 and 339.7 pence respectively, so:

*old average RPS = (271.3 + 274.8 + 339.7) / 3 = 295.3 pence*

1.2) Calculate average RPS for the new three-year period:

RPS for 2016, 2017 and 2018 was 561.6, 620.0 and 636.4 pence respectively, so:

*new average RPS = (561.6 + 620.0 + 636.4) / 3 = 606.0 pence*

1.3) Calculate revenue per share growth:

*RPS growth = (606.0 / 295.3) – 1 = 1.052 = 105.2%*

**2) Calculate capital employed per share growth:**

2.1) Calculate average CEPS for old three-year period:

CEPS for 2009, 2010 and 2011 was 177.1, 184.7 and 199.6 pence respectively, so:

*old average CEPS = (177.1 + 184.7 + 199.6) / 3 = 187.1p*

2.2) Calculate average CEPS for new three-year period:

CEPS for 2016, 2017 and 2018 was 361.0, 387.1 and 336.2 pence respectively, so:

*new average CEPS = (361.0 + 387.1 + 336.2) / 3 = 361.4p*

2.3) Calculate capital employed per share growth:

*CEPS growth = (361.4 / 187.1) – 1 = 0.932 = 93.2%*

**3) Calculate dividend per share growth:**

3.1) Calculate average DPS for old three-year period:

DPS for 2009, 2010 and 2011 was 12.0, 14.0 and 20.0 pence respectively, so:

*old average DPS = (12.0 + 14.0 + 20.0) / 3 = 15.3p*

3.2) Calculate average DPS for new three-year period:

DPS for 2016, 2017 and 2018 was 37.0, 38.9 and 41.3 pence respectively, so:

*new average DPS = (37.0 + 38.9 + 41.3) / 3 = 39.0p*

3.3) Calculate dividend per share growth:

*DPS growth = (39.0 / 15.3) – 1 = 1.548 = 154.8%*

**4) Calculate growth as the average of revenue, capital employed and dividend per share growth:**

*growth = (105.2% + 93.2% + 154.8%) / 3 = 117.7%*

**5) Calculate the growth rate:**

*growth rate = POWER(100% + 117.7%, 1/7) – 100% = 11.8% *

Burberry produced total overall growth for the ten years from 2009 to 2018 of 117.7%, which is an annualised growth rate of 11.8% per year.

This is well above my 2% rule of thumb minimum and well above inflation for the period as well. On that basis I think it’s reasonable to say that Burberry has an above average record of broadly-based growth.

Next week I’ll explain how I look for companies that have above average profitability as well as above average growth, because high profitability is one of the best indicators of a company with strong competitive advantages.

*Here’s a summary of the rules of thumb we’ve covered this week.*

Rules of thumb for consistent growth

Only invest in a company if its growth quality is above 75%.

Only invest in a company if its growth rate is above 2%.