Here’s what I do… I look at how the company handles depreciation and amortization of its assets. If management is trying to inflate earnings, it is highly likely that management will be aggressive in making aggressive estimated useful life assumptions behind D&A.
- It’s really obvious to the accountants that they can inflate earnings by making more aggressive assumptions here.
- Pulling this lever is perfectly legal since these assumptions are inherently uncertain and therefore subjective. No one will go to jail for pulling on this lever.
- All companies have to make assumptions on this accounting topic.
In most 10-Ks, there is a table that summarizes the main asset categories and the range of estimated useful life assumptions for that category. I generally try to find that table by searching for the phrases “straight-line” and “straight line” (if that doesn’t work, I try declining-balance).
Computer hardware generally follows Moore’s Law, decreasing in value every 2 years. This corresponds very closely to the declining-balance method with the computer hardware losing 30% of its value every year. After two years, 0.7 times 0.7 = 0.49 which is very close to losing half of the value after 2 years.
Another method for depreciating computer hardware is to use the straight-line method over 3 years. For the first year, earnings will be overstated. For the third year, earnings will be understated. If the company is rapidly growing, then this method will generally inflate earnings slightly as long as the company is growing.
If the computer depreciates computer hardware on a straight-line basis over 5 years, then earnings will generally be inflated. In my opinion, many tech companies like Amazon and Rackspace chose some straight-line method to give a small boost to their earnings. An honest person would likely use the declining-balance method as it closely models the true economic cost of depreciating computer hardware. (*If computer hardware is an immaterial portion of a company’s assets, then using straight-line for everything leads to simpler accounting. I am not bothered by non-technology companies that use straight-line for everything.)
One of the ways a company can egregiously game the accounting rules is to pretend that computer hardware is something else. Normally, tech companies will break out computer hardware (and software) into its own category. One of the ways company employees prepare financial disclosures is to look at what their peers are doing. Doing what everybody else is doing is generally a safe, cost-effective way of preparing financial statements. However, some tech companies will get “creative” and lump computer hardware (and software) into some other category. They can then use this as a smokescreen while they depreciate computer hardware over an unreasonably long period of time. So now you know how I feel about a certain data-center roll-up stock…
If a company depreciates a mine (the fixed assets tied directly to a mine) over a period longer than 20 years… it is a fairly obvious signal that the company has inflated books. See my post on mine economics.
My opinion is that most/all of them are assuming that their fixed assets will last much longer than they actually will. This includes Burlington Northern Santa Fe, the railroad that Berkshire Hathaway purchased.
In general, D&A at railroads is understated because the replacement cost of assets have gone up significantly due to the increase in commodity prices (e.g. steel). Their capex spending during recession years around 2009 should be a close proxy for maintenance capex (e.g. look at capex for all the years where rail volumes have declined). Overall, the management teams don’t seem particularly interested in helping investors actually understand the business and being honest with shareholders. These sorts of minor shenanigans generally occur at most publicly-traded companies. Institutional investors are partly to blame because they rarely understand what they invest in and focus too much on the short-term.
The accounting practice that I’m the most skeptical about is declining to provide any concrete information on the depreciation and amortization assumptions used. One railroad company does this because it doesn’t use the straight-line method for any of its assets (or the declining-balance method).
In many cases, it is difficult to tell what appropriate assumptions are. It can sometimes be helpful to find industry peers and see how they are handling their accounting. This may not be helpful for industries where everybody is engaging in shenanigans.
Some ways to figure out industry peers:
- Use a site like Google Finance, and look at the companies that are considered to be peers. The Google Finance data can be wonky sometimes.
- Read the DEF 14A and look at the companies in the peer comparison group. Companies often benchmark themselves against peer groups when determining compensation.
If the public-traded company has acquired other publicly-traded companies, it can be interesting to see the accounting for the subsidiary before and after the acquisition. For example, Ticketmaster’s accounting for computer hardware and software was more conservative before it was purchased by Live Nation.
“There’s never just one cockroach in the kitchen.”
If the company is playing accounting games in one area, it is likely that they are playing accounting games in other areas too. Usually this is the case, though not always.
Tracking changes in accounting assumptions
Tracking changes in the D&A assumptions is a good tool for figuring out whether the accounting has become more conservative or aggressive over time. For some asset classes like computer hardware, you can be fairly certain that the asset class has not changed materially year-over-year. Changes in the estimated useful lives of computer hardware suggests that management is up to something. The same applies for buildings, leasehold improvements, furniture, vehicles, etc.
This is a shortcut method
Keep in mind that this won’t always work. Sometimes the results are unclear. Sometimes the company has trained its investors to focus on some non-GAAP metric (e.g. adjusted cash EBITDA or whatever) and may try to deflate its GAAP earnings.