Message Font: Serif | Sans-Serif
No. of Recommendations: 24
We know Buffett focuses on ROE, but we rarely mention INCREMENTAL returns on equity.

For a refresher, here's an old article from MF. High incremental ROE makes me think of Walmart.

Incremental Return On Equity

By Maynard Paton (TMFMayn)
April 22, 2002

Warren Buffett once said:

"Leaving the question of price aside, the best business to own is one that over an extended period can employ large amounts of incremental capital at very high rates of return. The worst business to own is one that must, or will, do the opposite -- that is, consistently employ ever-greater amounts of capital at very low rates of return."

So how can investors detect a business that can employ "incremental capital at very high rates of return"? The trick is to use the incremental return on equity calculation, and not get misled by the more traditional return on equity ratio.

Like most things in investment, it's all best explained with an example.

My New Business -- the Early Years

Let's say I've invented a new type of radio and discover I can manufacture my new model far cheaper than anybody else. I go straight into business and start my company with £1,000, to buy all the necessary equipment. Shareholders' equity, reflecting all the assets the company has bought, is thus £1,000.

It's a great business. I make £300 post-tax profits in the first year. My return on the initial investment is 30%. In other words, the company's return on its starting equity is 30%.

Pleased with the first signs of success, I re-invest all the profits back into the business at the start of the following year. I buy more equipment to increase my manufacturing output. The new equipment increases the company's asset base, or equity, to £1,300.

The new £300 of equipment produces an additional £90 of profit over the second year, a return of 30%. Also, during the second year, my original £1,000 of equipment generates another £300 profit, as it did the year before. Overall, I've made £390 in my second year in business.

I repeat the whole process again in the following year. Here's a summary of my company's history after year three. For simplicity, I'll report the return on starting equity, rather than the more commonly used return on the average equity employed through the year.

Year Opening Profit Closing Return on
Equity (£) Equity opening Equity
(£) (£) (%)

1 1,000 300 1,300 30.0
2 1,300 390 1,690 30.0
3 1,690 507 2,197 30.0

Growing Pains

At the start of the fourth year, sales of my radios are slowing. So, I quickly launch a new television to boost profits instead. Profits are reinvested evenly into the existing radio business, and my new television business.

Thus, at the start of the fourth year, I plough £254 (half of year three's £507 profit) into each of my two businesses. So, my radio subsidiary has assets of £1,944, and my television subsidiary now starts with assets of £254, taking total company equity to £2,197.

But in my haste to release a new product, I've forgotten to do my homework on televisions. There's too much competition in this market, and my new product only generates returns of 5%. So year four profits of £596 comprise £583 (30% return on £1,944) through radios and £13 (5% return on £254) from televisions.

Undeterred, I evenly split my eventual profits into each subsidiary at the start of years five and six as well. Here's how things turn out.

Year Opening Equity Profit Closing ROE
Radio TV Total Radio TV Total Equity

4 1,944 254 2,197 583 13 596 2,793 27.1
5 2,241 551 2,793 672 28 700 3,493 25.1
6 2,591 901 3,493 777 45 822 4,315 23.5

Trouble Later On

The going gets really tough in year seven. A big competitor has set up in the radio industry and my profits stagnate, but the assets required to produce my radios still produce a 30% return. So, I concentrate heavily on the television business and my profits are diverted totally to reinvestment in this area.

So, the start of year seven sees me invest all the prior year's profit of £822 into the television subsidiary, boosting its asset base to from £901 to £1,724 I make a 5% return again from this equity, and another 30% from the now static radio asset base of £2,591. It gives me a total of £864 profit.

The same reinvestment strategy is made for the following three years. Here's how things turn out.

Year Opening Equity Profit Closing ROE
Radio TV Total Radio TV Total Equity

7 2,591 1,724 4,135 777 86 864 5,179 20.0
8 2,591 2,587 3,179 777 129 907 6,086 17.5
9 2,591 3,494 6,086 777 175 952 7,038 15.6
10 2,591 4,446 7,038 777 222 1,000 8,037 14.2

Review Of My Business Returns

As you're no doubt aware, I gradually moved from investing in a very high return business to investing into a very low return business. Was this apparent from the calculated return on equity figures?

If you were to look at my business in year five, the calculations for return on equity still look impressive. My overall profit then was £700 with £2,793 invested at the start of that year. My return on starting equity was still a high 25.1%. Moving on to year ten, the return was still a fair 14.2%.

The trouble with this return on equity calculation, dividing profit by the total equity accumulated in the business, is that the performance of "historic" investment decisions can mask the performance of more recent ventures.

Going back through this example, the original £1,000 radio investment was still producing a £300 annual profit at year ten. This investment contributed 30% to the final year's overall profit. That's all well and good, but it was obvious that these great returns of yesteryear couldn't be reproduced subsequently.

Incremental Return on Equity

What counts is how the business has invested its profits over the last few years. To identify whether recent reinvestment decisions are generating adequate returns, the incremental return on equity should be calculated using this formula, where x is the reinvestment period in years. For example, to calculate the return on incremental equity for the last 3 years, use x = 3.

Profit (year n) - Profit (year n - x)
Equity (end of year n) - Equity (end of year n - x)

The following table illustrates the incremental returns over one and three years for my business.

Year Return on Incremental Return on Equity
Opening Equity 1 year 3 Years
(%) (%) (%)

1 30.0
2 30.0 23.1
3 30.0 23.1
4 27.1 14.9 19.8
5 25.1 14.9 17.2
6 23.5 14.9 14.9
7 20.0 4.8 11.2
8 17.5 4.8 8.0
9 15.6 4.8 4.8
10 14.2 4.8 4.8

Note how the incremental returns fall much faster, and from a lower base, than the usual return on equity figures that appear to gently glide downwards. Indeed, the one year figures drop materially each time the proportion of profit going into each business alters.

The first three-year incremental figure that can be calculated, after year four, shows a result of under 20% when re-investment initially diverts away from the radio business. No doubt that calculation would raise an eyebrow to any shareholder who had got used to 30% returns at that point. The traditional return on equity figure at that time would have been a less understated 27.1%

Although the calculation slightly understates the true return, the incremental return on equity gives a great handle on the efficiency of recently reinvested profits. In summary, the incremental return on equity gives a far clearer and a very timely indication of possible profit re-investment underperformance.

Questions and comments to the Qualiport message board, please.

A version of this article was published in Janaury 2000
Print the post Back To Top
No. of Recommendations: 3
Another good one for the cognescenti . . . this one by Dale Wettlaufer: (scroll half-way down the page).
Print the post Back To Top
No. of Recommendations: 1
Hi Longreits,

Thank you for the nice example and I enjoyed reading through it.
I'd like to highlight that looking at the incremental return
on equity (by dividing recent profit change by recent equity change)
gives an accurate picture of how new capital is deployed in the
case that previously deployed capital is providing the same
return on equity. It doesn't work however when the return on
equity of earlier deployed capital is changing and this is commonly
the case.

Profit (year n) - Profit (year n - x)
Equity (end of year n) - Equity (end of year n - x)

If the recent profit (year n) in the formulae above increases
you still do not have any information about whether what
caused the profit to increase. It might have increased owing
to recently deployed capital however it might also have increased
(or decreased) owing to capital deployed ages ago starting to
produce more profits. In your example capital deployed was
usually assumed to continue producing the same return on capital,
so by looking at the difference in profits in recent years
with difference in equity you can accurately work out the return
on equity for the newly deployed capital.

To illustrate the problem, take Berkley Insurance:

Its share equity from 2007 to 2009 moved from 3569m to 3602m.
So share equity increased by 33m, while profits decreased.
Thus the incremental return on equity would be reported as
negative .. however new capital deployed might well have produced
positive returns on equity and just the bulk of the business'
profits decreased (associated with capital deployed much

As an opposite example, if your example, the original 1,000 pounds
deployed continued to produce the same profits. If the profits
associated with that investment had started to increase then the
incremental return on equity would be lifted higher (owing to
Profit (year n) rising compared to Profit (n - x)). So the
formulae would indicate an excellent incremental return on
equity in recent years although new investments might be terrible
and just earlier business was becoming more profitable.

Nevertheless your formulae above does leads one to investigate
what is actually happening in particular periods and I think is
a useful and interesting calculation to make.

- Manlobbi
Print the post Back To Top
No. of Recommendations: 0
Thanks, Manlobbi. When you're comparing ROEs from multiple years, do you use accounting net income, or do you use CFO or some other metric to more accurately gauge true cash flow?
Print the post Back To Top
No. of Recommendations: 1
One way I think about ROE:

ROE = net income/shareholder's equity

= (net income/sales)*(sales/assets)*(assets/equity)

= net profit margin * asset turnover * leverage

For KO, for example..

= 24% * 0.6 * 2.0

ROE = 30%

KO receives a modest boost from leverage, but the asset turnover is very low. Only 0.60 cents in sales for $1.00 in assets. Asset turnover and net profit margin move often in opposite directions ... high margin products (like KO's 24%) do not often accompany high sales per dollar of assets.

So, really, KO's economics turn on its huge profit margins - and what Buffett has called "untapped pricing power". They also turn on the ability to increase case volume indefinitely, even if slowly.

Conversely, PEP is running 17% net margins, asset turnover is 1.1 and leverage is 2.5.

PEP has lower margins but greater asset turnover and much more leverage. So, ROE is significantly boosted by the leverage.

I'm using net income here so we're talking about accounting earnings. Perhaps it would be better to use return on assets as a proxy for the net income margin and asset turnover, substituting in operating earnings less tax for "net income". Does using accounting earnings in computing ROE tell the whole story?
Print the post Back To Top