Let’s take depreciation. Say a company spent $X to put fibre in the ground. The company will now depreciate the fibre over some number of years, while selling fibre to customers and getting an income. Over time, the fibre may need replacement, but that replacement will essentially be a new investment. The depreciation is there to show the value of the assets but it’s not having a bearing on the cash other than requiring new investment some years down the road.
Anirban.
I disagree. If the company is in the fiber business replacement fiber is replacement, not new investment. But that is not really the issue.
When the company bought the fiber with an expected ten year useful life it figured that during each one of those ten years it would have some revenue that is not pure profit, you have to deduct some of the capital value of the fiber, in this case 10%, otherwise your profit calculations are all wrong.
Let’s take the case where a company invests in non-physical assets (computer code), calls it R&D, and expenses it rather than capitalizes it. When the code is being written you have a huge expense and little revenue from it. Later you have pure revenue with no cost. But if instead of computer code you had built a smelter you would capitalize it and depreciate it. (Very not the same for different companies. That GAAP standardizes accounting between companies is an urban legend). Let’s take an example of a software company, say Microsoft. Having expensed Windows the day it was written, Widows appears on the books at ZERO VALUE. No matter how crappy it might be, there is no doubt that Windows is valuable. Suppose for the sake of argument that Oracle bought Microsoft. It would pay a few million for Windows which is valued at zero. That excess payment becomes Good Will on Oracle’s books while it was non-existent on Microsoft’s books. How is it possible that the same “thing” is nothing on Microsoft’s books but it is an asset (goodwill) Oracle’s books? Not very standard reporting at all.
Expensing certain types of R&D is nothing but a tax gimmick, a loophole, industrial policy, or successful lobbying – take your pick. When Congress wants to give industry a helping hand it authorizes “accelerated depreciation.” Expensing certain types of R&D is “instant depreciation.” It’s not good accounting, but tax policy.
My point is this: Originally accounting was designed to help owners and mangers run their business. Then the tax man found it a useful tool to assess taxes but that creates a conflict of interest. If you want less taxes you hide profits. Then the joint stock company was invented and a new conflict of interest developed between shareholders and managers. Then Congress decided that fairness mattered so they put a straightjacket on the accountants. Now we have three sets of books, one for the taxman, one for owners (GAAP), and one for managers (ADJUSTED).
BTW, did Stratasys overpay? I would say “yes” if it paid cash. But what if it paid with overpriced stock? 
The bottom line is that you have to know your companies ONE BY ONE.
Denny (GAAP IS CRAP) Schlesinger
but beware “Adjusted CRAP!”
BTW, the fiscal year is totally arbitrary. Suppose the fiscal year were ten years and you installed the fiber in year one. At the end of the year it would be used up. No need to capitalize and depreciate, it would not affect your results. Depreciation is needed only when the useful life is longer than the fiscal year. Depreciation is an artifact to “synchronize” the “asynchronous” cash flow vs. cost.