The capability and appropriateness of measurement systems and related metrics are not just things that scientists and engineers must care about. This seems to be obvious to just about anyone, unless you were on Wall Street before the financial crisis.
The Deloitte Center for the Edge has published a report on the decline in the return on assets of American businesses over the past 40 years. Jon Taplin, a professor at USC, posted a very insightful summary of the report, likening this decline and how it was hidden to a shell game. What was interesting to me in his post was how the blind obedience to a particular metric has been in large part to blame for our current financial insanity.
What the Deloitte report points out is that companies have been able to “juice” their return-on-equity (ROE) numbers by consistently taking on more and more debt. Meanwhile, their return-on-assets (ROA) have fallen steadily. If you are even a casual investor or small businessperson, you’ve probably heard of ROE and why it is important. You may not have heard of ROA. Let me briefly explain the difference
Return on equity is a company’s annual net income divided by total shareholder equity. Shareholder equity is essentially how much money investors have put into your company, so ROE measures how effective you are at generating a return on invested funds. Return on assets, on the other hand, is your annual net income divided by total assets. ROA, therefore, measures your effectiveness at generating a return on everything the company owns and is in the bank.
You may already be seeing the disconnect, just based on my choice of words when I defined ROE and ROA above. Let me give you an example: Let’s say you have two companies, A & B. Each of these companies generates 1 M$ per year in net income. Each of these companies has 5 M$ in equity on the books, meaning that the investors have 5 M$ in them. In each case, the ROE of the company is 20%. Not too shabby. But there is an important difference between them. Company B also has 5 M$ of debt outstanding. Company B will thus have 5 M$ more assets on the books than Company A, and thus their ROA will be lower.
If you’re a CEO and you’re managed by your board on the basis of your ROE, you thus have a substantial incentive to leverage your company with loads of debt in order to have more resources with which to expand your business, since that debt doesn’t show up directly on your measurements. People will still invest in your company on the basis of your keen ROE (so long as they don’t look at your debt-to-equity ratio, or your actual return on assets.)
The bottom line here is that a lot of people had a warning right in front of them about what was happening with GM and other companies, but couldn’t see it because one of their chief metrics hid it from them. As with so many other things, relying on a few simple metrics is dangerous and sloppy. Simple metrics are useful, but they must be cross-checked and reviewed with a constant eye on exactly what they tell you and what they do not.