Last week, a reader responded to a post discussing hhgregg's (HGG) ROE breakdown with the statement that "I'm not sure if ROE is the right metric" and subsequently provided some calculations whereby he concluded that he believes "it's clear that this company does not manage to earn its cost of capital especially if we assume a mild 12% cost of equity."
The reader is correct that ROE is not always the best metric; but whether it's the best metric to use depends on what is being measured. If one is trying to measure the rate of return that has been achieved on the firm's equity investments, including retained earnings, there is no better measure. On the other hand, if one would like to know whether the firm achieves a good rate of return on all of its invested capital (i.e. equity and debt combined), ROIC would be better.
Some would argue that shareholders should care mostly about ROE, since shareholders have claim on only the equity returns. But returns on equity can be risky. If a lot of debt is used to finance a business, fixed costs and obligations are high, leaving the firm vulnerable if trouble hits. For this reason, knowledge of both ROIC and ROE are necessary to understand a business.
To that end, the reader provides some rough calculations for HGG's most recent year's ROIC. Unfortunately, there are different formulas for this calculation, as different people have different ideas about exactly what should be considered invested capital. While I agree with the reader that off-balance sheet leases should be included as part of capital, we disagree about whether certain liabilities should be included as invested capital (see our comments in the post for more details). These disagreements end up having a large effect on the result. Simple differences in assumptions result in ROIC discrepancies of 150 basis points (between 8.7% and 10.2%).
Armed with the results of his ROIC calculation, the reader then compares HGG's ROIC to a cost of equity of 12%. Unfortunately, this is not an apples to apples comparison; it is ROE that should be compared to the cost of equity, while ROIC should be compared to the cost of capital. This is because the firm generates ROIC at the cost of both debt and equity, whereas ROE is generated at just the cost of equity. Because the cost of debt is lower than the cost of equity, and because debt has a tax shield, the firm's cost of capital is much lower than its cost of equity.
My calculations show that hhgregg is adding value because its ROIC is higher than its cost of capital. Obviously, however, different assumptions will yield different results. Readers should therefore not put too much faith in any one number.
Finally, whether using an ROIC or ROE metric, it's important to look further than just one year's worth of data. The consumer is going through a tough period right now, and recent sales of items that have some correlation with the housing market are abnormally depressed. What's important is how a company deals with changing preferences over time, not whether they were caught surprised in a given quarter that then had an effect on the most recent year.
Also in hhgregg's favor is the fact that, as a growing company, it is experiencing costs for revenues it has not yet received. As the company builds out distribution centers and trains staff for its new locations, it's spending money without yet enjoying the benefits.
So what's the conclusion? It depends. Both ROIC and ROE are useful and important. Investors should look at both and rely on neither.