Edit: If you buy miners using your company, and you don't expense them right away, I would suggest the double declining method for amortizing your equipment costs if you're worried about difficulty increases and profits diminishing over time, which I also agree is pretty much a given at this point.
Hmmm, that's a hard one, you might be right
You certainly need an accelerated depreciation method simply because mining time lost today is worse than mining time lost tomorrow (because of the increasing total network hashrate & following difficulty).
Normally the double declining balance method assumes faster depreciation in the beginning of the asset's life and slower depreciation near the end to account for lower maintenance costs in the beginning of the asset's life vs. the end, but it could be seen to account for the above as well.
I think the only hope we have is that Moore's law continues to prove true for mining equipment. If hashrates and difficulty is curvilinear, we know for sure that so is processing power according to Moore's law. So, I will say this: it is no surprise that this far into the product life cycle of the S9 we are looking at these types of predictions.
Although the expectancy for sustainable levels of profitability is still debatable, what is not debatable is that we will sooner or later hit the point where the S9 generates less BTC (at that moment's BTC/USD rate) than it uses in electricity (at that moment's USD price per kWh). I believe the point that is trying to made is that if difficulty rises and we hit a point where you spend $300 in electricity per month, but you only mine .025 BTC per month, you are paying the equivalent of $12,000 per BTC. And if the USD/BTC exchange rate ends up being $11,000, then it doesn't make much sense to keep mining at that price because if you are counting on appreciating of value, you get more buying 1 BTC at the $11,000 rate than .025 at the $12,000 rate (your electricity costs vs the amount of BTC mined).
Hope that doesn't confuse anybody too much. Let me put this in perspective: at the current difficulty I am currently scoring about .06 BTC per month with my S9 and paying $150 for the electricity to keep it running. That means the S9 is allowing me to gain BTC at a $2,500 purchasing price. This is a great deal if the USD/BTC rate stays at $11,000 (Profit ratios are at 440%). If difficulty rises to where my S9 only generates .01 BTC per month for the same $150 in electricity costs, that same S9 is now allowing me to gain BTC at a $15,000 purchasing price. This is a terrible deal if the USD/BTC rate stays at $11,000 but if the BTC/USD rate jumps to $25,000 its a great deal again (Profit rations are at 167%).
The only thing this doesn't take into consideration is the purchase price of the equipment and its salvage value at the end of its life, notwithstanding malfunctions or dead equipment.
Dont worry price will go up, lightning network is coming soon that will kill a bunch of altcoins and make bitcoin what it should be painless, fast way to transfer money not just a way to store wealth and speculate
In the case mining gets too hard before they release new miner a lot of hashing power is gonna turn off doubt somebody is gonna mine losing money
Just keep mining and storing that BTC until next december ull see...