I would not call $15k a touch below - which it currently is. I've been watching the price difference the past 6 months and the cost to mine 1 BTC has been anywhere from $10,000 - $30,000 more than buying 1 BTC during that period.
Miners are buying BTC because they are losing mining it. But without mining BTC stops working because transactions need to be confirmed. Bitcoin will never stop as long as there are individuals willing to mine it. But what will happen is mining will centralize more making Bitcoin's security model centralized. It's also vulnerable to energy supply issues because so much energy is required to maintain the hash rate.
For now, this won't be an issue. But I see it becoming a problem 16-24 years from now when miner yields will be 1/16th to 1/64th what they are now. Transaction fees would have to make up the difference to keep the network security high. But if BTC is a store of value and you're supposed to hold it, where will these transactions come from?
First of all, there are different published stats of what it costs to mine. The reality is, no knows for certain what the true average cost is, although we can make approximations.
Second, some starts show the average cost as being just marginally over market value, as in 1-5K (not 15K).
Third, what we do know, is that over time, the cost of production and market price have always been correlated (sometime over, sometimes under, for obvious reasons), but over time they converge and revert around each other. So, for example, let's assume your average cost of 80K per coin is correct - we could easily have a jump to BTC to 85-100K, at which point, mining would be highly profitable. And then after a period of time, hash rate goes up, mining gets less profitable, and cost of mining reverts again to the mean, and maybe even drops below due to optimistic over-investment in mining rigs. And then cycle repeats.
Fourth, with an "average cost", we know some miners will be mining cheaper than that average and some more expensive. So, those with the best equipment, and access to cheapest energy will be below that average, and some of course will be well below.
Fifth, many miners have an incentive to inflate cost of mining they report. Reasons include to maximize the reported total production cost to maximize tax losses, putting almost any expense of the company (including long term capital investment, advertising, cost of labour, lavish travel etc into the "cost", which if really needed they could easily cut.
Sixth, many miners report an inflated cost so as to deter competitors from entering. So, let's say you are making 10K a coin. What you will be telling the market is that even with your highly efficient operation, you are losing 15K a coin. This helps deter new entrants, and helps protect margins.
Seventh, the reason miners are buying coins, is fundamentally, because they consider its a better action that selling coins or retaining cash. They continue to mine like crazy and on top they hodl. These miners now remaining really know their stuff, and are making far more mathematically sophisticated decisions, hedging, arbitrage, forecasting and financial modeling etc.
Eighth - mining will wane thorough various phases of centralization, and decentralization, due to numerous economic, political, technological, and free-market factors. We will also see a "decentralisation within centralization" framework emerge over time. I will write a longer post on this at some point, but I think you and Mayday may already have a sense of what I mean here.
OK - signing off for now. Let the BTC pump strong! (All other alts will follow - but don't forget, the supply of alts is infinite, where the total supply of global wealth is not).
