Accounting Returns, Market Returns and Good/Bad Businesses
Back to Basics
While I was working on my last two data updates for 2025, I got sidetracked, as I am wont to do, by two events. The first was the response that I received to my last data update, where I looked at the profitability of businesses, and specifically at how a comparison of accounting returns on equity (capital) to costs of equity (capital) can yield a measure of excess returns. The second was a comment that I made on a LinkedIn post that had built on my implied equity premium approach to the Indian market but had run into a roadblock because of an assumption that, in an efficient market, the return on equity would equate to the cost of equity. I pointed to the flaw in the logic, but the comments thereafter suggested such deep confusion about what returns on equity or capital measure, and what comprises an efficient market, that I think it does make sense to go back to basics and see if some of the confusion can be cleared up.
The Lead In: Business Formation
To keep this example as stripped of complexity as I can, at least to begin, I will start with two entrepreneurs who invest $60 million apiece to start new businesses, albeit with very different economics:
The first entrepreneur starts business A, with a $60 million investment up front, and that business is expected to generate $15 million in net income every year in perpetuity.
The second entrepreneur starts business B, again with a $60 million investment up front, and that investment is expected to generate $3 million every year in perpetuity.
With these characteristics, the accounting balance sheets for these companies will be identical right after they start up, and the book value of equity will be $60 million in each company.
The return on equity is an entirely accounting concept, and it can be computed by dividing the net income of each of the two businesses by the book value of equity:
Return on equity for Business A = Net income for Business A / Book Value of Equity for Business A = 15/60 = 25%
Return on equity for Business B = Net income Business B / Book Value of EquityBusiness B = 3/60 = 5%
Assume that both these businesses have the same underlying business risk that translates into a cost of equity of 10%, giving the two businesses the following excess returns:
Excess Return for Business A = Return on equity for Business A – Cost of equity for Business A = 25% -10% = 15%
Excess Return for Business B = Return on equity for Business B – Cost of equity for Business B = 5% -10% = -5%
In the language of my last post, the first business is a good one, because it creates value by earning more than your money would have earned elsewhere on an investment of equivalent risk, and the second is a bad one, because it does not.
The return on equity may be an equation that comes from accounting statements, but in keeping with my argument that every number needs a narrative, each of these numbers has a narrative, often left implicit, that should be made explicit.
On business A, the story has to be one of strong barriers to entry that allow it to sustain its excess returns in perpetuity, and those could include anything from a superlative brand name to patent protection to exclusive access to a natural resource. In the absence of these competitive advantages, these excess returns would have faded very quickly over time.
On business B, you have a challenge, since it does seem irrational that an entrepreneur would enter a bad business, and while that irrationality cannot be ruled out (perhaps the entrepreneur thinks that earning any profit makes for a good business), the reality is that outside events can wreak havoc on the bet paid plans of businesses. For instance, it is possible that the entrepreneur’s initial expectations were that he or she would earn much more than 5%, but a competitor launching a much better product or a regulatory change could have changed those expectations.
In sum, the return on equity and its more expansive variant, the return on invested capital, measure what a company is making on the capital it has invested in business, and is a measure of business quality.
The Market Launch
Assume now that the owners of both businesses (A and B) list their businesses in the market, disclosing what they expect to generate as net income in perpetuity. Investors in equity markets will now get a chance to price the two companies, and if markets are efficient, they will arrive at the following:
Thus, a discerning (efficient) market would value business A, with $15 million in net income in perpetuity at $150 million, while valuing business B, with $3 million in net income in perpetuity, at $30 million. If you are wondering why you would discount net income, rather than cash flow, the unique features of these investments (constant net income, no growth and forever lives) makes net income equal to cash flow.
Even with this very simplistic example, there are useful implications. The first is that if markets are efficient, the price to book ratios will reflect the quality of these companies. In this example, for instance, business A, with a market value of equity of $150 million and a book value of equity of $60 million, will trade at 2.50 times book value, whereas company B with a market value of equity of $30 million and a book value of equity of $60 million will trade at half of book value. Both companies would be fairly valued, though the first trades at well above book value and the second at well below, thus explaining why a lazy variant of value investing, built almost entirely on buying stocks that trade at low price to book ratio,, will lead you to holding bad businesses, not undervalued ones.
As I noted at the start of this post, it was motivated by trying to clear up a fundamental misunderstanding of what return on equity measures. In fact, the working definition that some commenters used for return on equity was obtained by dividing the net income by the market value of equity. That is not return on equity, but an earnings to price ratio, i.e., the earnings yield, and in these examples, with no growth and perpetual (constant) net income, that earnings yield will be equal to the cost of equity in an efficient market.
Extending the Discussion
One of the advantages of this very simple illustration is that it now can be used as a launching pad for casting light on some of the most interesting questions in investing:
Good companies versus Good Investments: I have written about the contrast between a good company and a good investment, and this example provides an easy way to illustrate the difference. Looking at companies A and B, there is absolutely no debating the fact that company A is better company, with sustainable moats and high returns on equity (25%), than company B, which struggles to make money (return on equity of 5%), and clearly is in a bad business. However, which of these two companies is the better investment rests entirely on how the market prices them:
As you can see, the good company (A) can be a good, bad or neutral investment, depending on whether its is priced at less than, greater than or equal to its fair value ($150 million) and the same can be said about the bad company (B), with the price relative to its fair value ($30 million). At fair value, both become neutral investments, generating returns to shareholders that match their cost of equity.
The Weakest Link in Excess Returns: The excess return is computed as the difference between return on equity and the cost of equity, and while it is true that different risk and return models and differences in risk parameters (relative risk measures and equity risk premiums) can cause variations in cost of equity calculations, the return on equity is the weaker link in this comparison. To understand some of the ways the return on equity can be skewed, consider the following variants on the simple example in this case:
Accounting inconsistencies: As an entirely accounting number, the return on equity is exposed to accounting inconsistencies and miscategorization. To illustrate with our simple example, assume that half the money invested in business A is in R&D, which accountants expense, instead of capitalizing. That business will report a loss of $15 million (with the R&D expense of $30 million more than wiping out the profit of $15 million) in the first year on book capital of $30 million (the portion of the capital invested that is not R&D), but in the years following, it will report a return on capital of 50.00% (since net income will revert back to $15 million, and equity will stay at $30 million). Carrying this through to the real world, you should not be surprised to see technology and pharmaceutical companies, the two biggest spenders on R&D, report much higher accounting returns than they are actually earning on their investments..
Aging assets: In our example, we looked at firms an instant after the upfront investment was made, when the book value of investment measures what was paid for the assets acquired. As assets age, two tensions appear that can throw off book value, the first being inflation, which if not adjusted for, will result in the book value being understated, and accounting returns overstated. The other is accounting depreciation, which often has little to do with economic depreciation (value lost from aging), and subject to gaming. Extrapolating, projects and companies with older assets will tend to have overstated accounting returns, as inflation and depreciation lay waste to book values. In fact, with an aging company, and adding in stock buybacks, the book value of equity can become negative (and is negative for about 10% of the companies in my company data sample).
Fair Value Accounting: For the last few decades, the notion of fair value accounting has been a fever dream for accounting rule writers, and those rules, albeit in patchwork form, have found their way into corporate balance sheets. In my view, fair value accounting is pointless, and I can use my simple example to illustrate why. If you marked the assets of both company A and company B to market, you would end with book values of $150 million and $30 million for the two companies and returns on equity of 10% for both firms. In short, if fair value accounting does what it is supposed to do, every firm in the market will earn a return on equity (capital) equal to the cost of equity (capital), rendering it useless as a metric for separating good and bad businesses. If fair value accounting fails at what it is supposed to do, which is the more likely scenario, you will end up with book values of equity that measure neither original capital invested nor current market value, and returns on equity and capital that become noise.
Growth enters the equation: For companies A and B, in this example, we assumed that the net income was constant, i.e., there is no growth. Introducing growth into the equation changes none of the conclusions that we have drawn so far, but it makes reading both the return on equity and the earnings yield much messier. To see why, assume that company A in the example continues to have no growth, but company B expects to see compounded annual growth of 50% a year in its net income of $3 million for the next decade. We can no longer consign company B to the bad business pile as easily, and the current earnings to price ratio for that company will no longer be equal to the cost of equity, even if markets are efficient. Incorporating growth into the analysis will also mean that net income is not equal to cash flow, since some or a large portion of that net income will have to get reinvested back to deliver the growth. In fact, this is the argument that I used in my second data update to explain why comparing the earnings yield to the treasury bond rate is unlikely to yield a complete assessment of whether stocks are under or over valued, since it ignores growth and reinvestment entirely.
Exiting bad businesses: This example also helps to bring home why it is so difficult for companies in bad businesses to fix their "badness" or exit their businesses. In the case of company B, for instance, telling the manager to find projects that earn more than 10% is advice that can be freely dished out, but how exactly do you invent good projects in a business that has turned bad? While exiting the business seems to be a better choice, that presupposes that you will get your capital ($60 million) back when you do, but in the real world, potential buyers will discount that value. In fact, if you divest or sell the bad business for less than $30 million, you are actually worse off than staying in the business and continuing to generate $3 million a year in perpetuity, which has a $30 million value. In the real world, most companies in bad businesses hire new CEOs, restructure their businesses and enter new businesses in a desperate attempt to become good businesses, and enrich consultants and bankers, but not their own shareholders, along the way.
Conclusion
Many of the comments on my seventh data update, and on my explanation about why ROE and cost of equity don’t have to be equal in an efficient market, came from people with degrees and certifications in finance, and quite a few of the commenters had “finance professional” listed in their profile. Rather than take issue with them, I would argue that this misunderstanding of basics is a damning indictment of how these concepts and topics are taught in the classroom, and since I may very well be one of the culprits, one reason that I wrote this post is to remind myself that I have to revisit the basics, before making ambitious leaps into corporate financial analysis and valuation. For those of you who are not finance professionals, but rely on them for advice, I hope this is a cautionary note on taking these professionals (consultants, appraisers, bankers) at their word. Some of them throw buzzwords and metrics around, with little understanding of what they mean and how they are related, and it is caveat emptor.
YouTube Video
Thank you, it was very interesting!
I can't get enough of these. so fundamental and explained so well.