What Makes for Good Business? Part 2: Beating Your ROIC Benchmark

In our previous post, we spent some time on accounting fundamentals, trying to get a sense of the building blocks to measure a business’ profitability, uses of capital, and efficiency at which they can employ that capital. We tied it together with our metric of return on invested capital (ROIC), essentially our after-tax operating profit of the business divided by the net operating assets or invested capital. There are a number of different accounting treatments to come up with these numbers—for our purposes here, it’s more important to understand the general idea. As you recall, we determined that Kroger Co. earned a 10% ROIC in 2022.

To take it a step further, we need to put this into proper context. Who is to say that 10% is a good or bad ROIC for Kroger? We can look at historical performance over several years, but this only gives us whether 2022 was better/worse than average for Kroger. We can compare to other firms, many of which will likely earn higher returns on capital—software firms, for instance, require less incremental capital in order to grow, and returns can be much higher. But we still don’t have any benchmark to decide where a decent return becomes a bad one—is it 10%? 5%? Zero?

To answer this, we need to look at a couple theoretical concepts.

Capital—It’s Called Capital-ism For a Reason

Just like its counterparts, capitalism has been given many definitions by both proponents and opponents. But at a basic level, it’s in the name—essentially about the allocation of wealth, or capital. Suppliers of funds (people with money) and demanders of funds (people who want money to grow their businesses) operate in financial markets to determine who gets money to do what. When you borrow money to buy a home, you’re on the demand side of this giant market. When you put money in your bank account, buy/sell investment securities, or lend money to friends, you’re on the supply side.

This capital isn’t free. You may lend your buddy $5 at zero interest because you know he’ll pay you back, but suppliers of funds will insist on having a good reason to put their money in someone else’s hands. In most cases, this reason is the promise and expectation of a return on that money.

Most businesses (like Kroger) operate on the demand side of this capital market; they employ funds from a variety of sources to set up and operate their business (some financial companies operate as both employers and suppliers of capital, creating a bit more complexity). Companies can source this capital in two principal ways—debt and equity.

Debt is relatively simple, and something most of us are likely familiar with. A company borrows money, whether from a bank or from the investing public, and ultimately repays the amount they’ve borrowed, along with interest. This interest can either be paid back periodically throughout the life of the loan, or all at the end in one lump sum—or a combination of the two. The rate of interest a company pays is determined by several factors: length of the loan’s life, prevailing market interest rates at time of borrowing, and creditworthiness of the company are among the most important factors.

Investors in securities like bonds are essentially lenders to the company. They’re entitled to the interest payments and the principal repayment, both of which are components of the investor’s realized yield, or rate of return, on their investment. This makes estimating a company’s cost of debt fairly simple. The effective cost of current debt is simply the company’s interest expense divided by the average debt over the period (under current accounting standards, we have to make sure we’re dealing with leases correctly—but this must wait until another time).

If we want to get a forward-looking view, we need to consider the marginal cost of debt: the interest rate on whatever future borrowings the company may draw on. This requires accepting some uncertainty, but the same critical factors—prevailing interest rates, credit quality of the company, etc.—will be paramount.

Equity is where things get more difficult. Holding equity in the company entitles investors to ownership of the company proportionate to the equity (usually shares of stock) they hold. Just like the bondholders, they have expectations of getting return on their investments—whether in form of dividends or share price appreciation. Funding a business with equity is essentially using the owner’s money, but this can take two different forms: internal and external.

Internal equity funding comes from the money a business already has—money which in principle belongs to the existing shareholders/owners of the business. This is the simplest way for businesses to fund their growth. It also provides an indirect clue about management’s view of the company’s cost of capital, since management will almost always have more information about the company than anyone else—but this is difficult to quantify.

External equity funding comes from selling additional stakes in the business. This is most common in IPOs, when a previously private company offers its shares to the public, both to fund growth and allow founders with concentrated ownership to monetize some of their ownership. For established companies, external equity funding can create grumbling among existing shareholders—who are seeing their stakes in the company become diluted as new shares are sold.

Equity in general is more complex because the return an investor receives isn’t fixed or promised—it’s a claim on tangible business results that are highly variable. Because of this, cost of equity is better thought of as a concept rather than a calculation, because it depends heavily on the expectations and opportunity costs of individual investors. If a company’s earnings per share (EPS) increases by 5%, existing shareholders are all 5% richer. But if I’m expecting a 10% return on my investment (possible because I felt like that’s what I would have earned had I invested in a different company), I will likely be disappointed.

In this case, my expected return of 10% is part of the company’s cost of equity. It follows naturally, of course, that companies that fail to cover their costs of equity will ultimately become spurned by investors who are seeking better returns—this will make the company’s shares cheap, and this cheapness will by definition boost returns that investors can expect by buying the stock, since the underlying business remains the same. Because of this effect, in the very long run, all companies should earn their cost of capital, and nothing more—well-performing companies become more expensive, and weaker companies become cheap. Anomalies can and do persist for decades, but nothing—and I mean nothing—can last forever.

CAPM and WACC: Calculating the Incalculable?

This fundamental slipperiness of cost of equity has not deterred a few notable attempts to boil it down to a formula. The most well-known example of this is the “Capital Asset Pricing Model”, or CAPM. While indispensable in navigating through business school, the CAPM, in my view, makes some problematic assumptions that leave it better confined to academia. I’ll give an overview of the formula and my grievances, and then we’ll move on.

The CAPM breaks an asset’s expected return (expected return can be thought of in our context as another name for cost of equity) into two components—the risk-free rate, and the premium (difference) between the equity market return and the risk-free rate, scaled to the relative risk of this particular stock. The formula is:

Expected Return = Risk-free rate + Beta x Equity Risk Premium

To use this formula, you take an observable rate you assume to be risk-free (most often a U.S. Treasury bond), and add the premium on the equity market over the long-term (basically, the average amount by which the equity market outperforms the risk-free asset you’ve picked) multiplied by the company’s beta, or relative “risk” vs. the market. One thing this approach does well is result in a changing cost of equity based on market conditions: if interest rates are higher, you can earn more from Treasuries and should expect more compensatory return to hold equities instead.

Risk is a slippery term to define, but the CAPM (and academic finance in general) considers price volatility to be the best proxy for risk. Therefore, beta is a measure of an asset’s price volatility relative to that of the market. The market has a beta of 1.00. Higher betas are more volatile, lower betas are less volatile. One way to think about it is this—a stock with a beta of 1.5 is likely to go up 15% when the market goes up 10%, or go down 15% when the market is down 10%. We can estimate the beta of any stock by using some regression math that gets complicated until we put it in Excel.

Applying this to Kroger would look something like the following. Take the 10-year Treasury yield at time of this writing, which is 4%. A common equity risk premium assumption is 7%, so we’ll stick with that. And Kroger’s beta over the last three years is 0.28 (relatively low-risk, which makes sense given their business). This gives a cost of equity of 5.96% for Kroger.

If we were formula addicts, we could plug this resulting cost of equity into another formula—the Weighted Average Cost of Capital, or WACC—to incorporate the cost of debt, cost of equity, and the resulting total cost of capital for the firm. This formula should make intuitive sense:

WACC = Weight of Debt x Cost of Debt x (1 – tax rate) + Weighted of Equity x Cost of Equity

In this case, the weights should sum to 100%, and are the proportions of debt & equity in the company’s capital structure. We adjust the debt side for a tax effect, since in most cases interest expenses are tax deductible, and therefore add an advantage of debt financing.

We won’t bother with running through the math for Kroger on this front, because I would contend that, alas, our cost of equity number is not reliable. Here are a few concerns I have about the CAPM approach:

• The risk-free rate is never risk-free. Sure, the United States and other major economies have avoided defaulting on their debt—but the U.S. came awfully close in 2023. Even if you make the case that the current monetary system makes it impossible for America to default because of its reserve currency position, that position is not guaranteed, and many once-thriving nations have suffered from government financial mismanagement (see: Argentina). Even more problematic, interest rates can differ, even between countries of similar risk levels (U.S. and Japan are one historical example).
• Of course, the equity risk premium is sensitive to measurement periods (do you go back 10 years, 20 years, or 100 years?), but it can also be difficult to define the “market”. Does that mean the S&P 500—the index including the 500 largest U.S. companies? Should it include international stocks or small-cap U.S. stocks? Finding a reliable, all-encompassing index is important if your opportunity set is truly global, but this is easier in theory than in practice.
• Risk is NOT simply price volatility. This line of thinking if like buying home insurance in case of an earthquake, but also buying home insurance to protect yourself from your house doubling in value. When volatility works in our favor as investors, it’s foolish to consider that as risk—it’s opportunity. By equating risk and volatility, the CAPM is recommending you demand extra compensation in the event your investment performs so well that you get surprised. Where does that leave us? Risk is ONLY about the chance of you losing your money. Plus, the non-normal distribution of most risks doesn’t allow for it to be easily reduced to a standard deviation.

I find WACC a more robust concept than the CAPM, but this still comes with a quick caveat. In general, the cost of debt will be lower than the cost of equity (especially after-tax). Because of this, using more and more debt in the capital structure can give the company appearance of cheaper and cheaper capital—if the cost of equity is held constant. But cost of equity moves in response to higher leverage. Remember, equity holders own the company, but their claim’s value depends on debt being paid. So, if a company levers up with more debt, equity holders will see it as riskier, and demand a higher return. For each company (in theory), there’s an optimal point at which the cheap-debt and financial distress effects offset, to create an “optimal” capital structure at the minimum WACC—this theory was first introduced by Modigliani and Miller in the 1960s. 

Here are a couple alternative ways to get a cost of equity number.

1: Implied cost of equity. We always know the market price. If we take some forward estimate of the company’s earnings (whether from Wall Street or the company’s own guidance), we can use that to back out the cost of equity that the current stock price is implying. The only hiccup here is that we need some longer-term forward growth rate to earnings. We can make a rough estimate here, rely on sell-side estimates if we have access to them, or use any long-term projections the company provides—but these are often aspirational. We can then use the discounted growth formula in the following way:

Price = Next Year’s Earnings Estimate / (cost of equity – long-term growth).

Solve for cost of equity, and that’s approximately what the market is implying. The benefit of this is that it’s observable and reflects actual buyers and sellers of the stock, giving a true market-based approach. The downside is that it’s making a lot of assumptions. It’s very sensitive to whatever growth rate you assume. Plus, the buy-side may not be taking Wall Street estimates at face value—meaning you’re applying sell-side assumptions to try to see what the buy-side is thinking.

2: Just pick a number. While certainly lacking in mathematical rigor, this approach insists we accept what we can’t know for sure. CAPM or implied approaches are precise, but raise questions about consistency and accuracy. Simply picking a cost of equity number ourselves makes some sense simply because we are the prospective equity holders. Whatever our required return is, that should be how we make decisions—whether you use 10%, 12%, or something else is ultimately up to you. I would argue this approach is more appropriate when investing for the longer term. In shorter time horizons, stock prices can be dominated by market responses to cost of capital changes—Federal Reserve rate actions being the most notable example. If frequent buying and selling is part of your strategy, you don’t have the luxury of being agnostic about that. But for long-term buying and holding, you could just use an “average” estimate of cost of equity of your choosing. This feels lazy because it’s not as quantitative, but it shouldn’t be. Understanding average costs of equity across different markets, geographies, cycles, and industries takes experience and wisdom well-applied.

What does this all mean, anyway?

The point of estimating cost of capital is comparing it to our estimates of return on capital. We already suggested that in the very long run, businesses will earn their cost of capital—that is, true economic profit will be zero. But outside such an eternal view, businesses will often out-earn or fall short of their cost of capital, generating positive or negative economic profits. In general, any business that earns a better return than its cost of capital is earning a good return, since their business performance is creating more value for owners and creditors than those investors demand. Likewise, any company that earns less than its cost of capital is destroying investor value—even if the company is profitable—because investors are forgoing higher returns elsewhere, or simply aren’t being compensated enough for the company’s level of risk.

Economic profit is basically the “spread” between return on and cost of capital, multiplied by the amount of capital. If Kroger’s cost of capital is below our 10% ROIC, they will be earning economic profit.

We’ve covered two different cost of capital measures: cost of equity, and WACC (which includes cost of equity). The return measure and the capital cost measure we use need to match up. If we’re using ROIC, which is pure operating return on all capital (regardless of debt/equity source), we should use the WACC cost of capital metric, to capture the impact of this leverage on their cost of capital. If we use return on equity (ROE) as our return metric, we should compare it to the cost of equity, since ROE only captures net income (or equity earnings) and equity value (excluding debt capital). Again, I’m standing on the consistency soap box here, but keeping things apples-to-apples is critical to avoid making silly mistakes (of which I’ve made plenty) when evaluating a business. Now we’ve got some understanding of what components go into ROIC, how accounting numbers give clues about business performance, and what sorts of cost of capital benchmarks we can apply to determine what returns are good or bad. In our next post, we’ll introduce the concept of competitive advantage to try and understand how the anomaly of non-zero economic profits can come to be, and companies seek to sustain and strengthen their positions.

---