4 ms·
Their timeline is "the next decade" ... by that time, Bitcoin will have proceeded two more halvings, which means the issuance rate will be only 1.5625BTC per bl
by nnx 6y ago
Their timeline is "the next decade" ... by that time, Bitcoin will have proceeded two more halvings, which means the issuance rate will be only 1.5625BTC per block, so only about $40B per year is needed to reach equilibria.
Regarding inflation, the Fed has just confirmed yesterday their intent to increase it, they will for the first time not fight against inflation going more than 2% yearly over a period of time even...
All this money printing will end up in inflation at one point, which is also the only way to ever repay all that debt.
- arcticbull 6y ago> Regarding inflation, the Fed has just confirmed yesterday their intent to increase it, they will for the first time not fight against inflation going more than 2% yearly over a period of time even... Their issue right now is they're tracking towards 0.44%, not 2% as they intend. Their goal is to increase inflation towards 2%. > All this money printing will end up in inflation at one point, which is also the only way to ever repay all that debt. This is a common misconception among armchair economists (and a great point against Bitcoin, too). The issue we're facing right now is that the velocity of money has been reduced, causing deflationary pressure (not inflation). This is typical in recessionary environments, and not at all desirable. Satoshi's will notwithstanding. Further, as long as the expansion of the broader economy outpaces the growth in deficit it doesn't actually need to be paid back. Also, more than half of US debt is domestically held.
- ro-_-b 6y agoI believe it is not fair to only take into account consumer price inflation instead of asset price inflation. Asset price inflation in the US & Europe has been massive lately. BTC is obviously an asset and therefore the inflation that is relevant for the future valuation of an asset IMO is asset price inflation
- arcticbull 6y agoAsset price increase is not inflation, which is literally defined relative to CPI. If we're going to talk about economics we should use broadly accepted definitions yeah? > BTC is obviously an asset... Hold on now. Nobody seems to agree on that.
- ro-_-b 6y agoif you look at how expensive gold is you would typically compare it in relation to other assets such as real estate or the S&P 500 and not to CPI. that's at least how I would make up my mind when is a good time to buy sell/gold. Gold/precious metals are an asset class and BTC/crypto-assets are an emerging asset class IMO also I think it is not fair that central banks only focus on CPI and not asset price. for me as a millennial it matters a lot that real estate price in central Europe are today 2x compared to what they were 10 years ago. also rent is mostly not reflected correctly in CPIs
- arcticbull 6y ago> IMO also I think it is not fair that central banks only focus on CPI and not asset price. I can't figure out why it's not fair, though. All that matters with asset prices is whether they go up or down, not their absolute value. The whole point of assets is they go up. Your complaint is they, what, went up "too fast" or that you weren't in when they did? > ...for me as a millennial it matters a lot that real estate price in central Europe are today 2x compared to what they were 10 years ago. Well, inflation accounts for I suspect a whole lot of that. Inflation since 2010 is just shy of 20% for the USD. I'm unfamiliar with central European pricing so I can't comment on that specifically, however it might surprise you to know that in the US, on an inflation-adjusted dollars per square foot basis, houses cost exactly the same as they did in the 1970s. They're just twice as big now. [1] > ...also rent is mostly not reflected correctly in CPIs. In what way? [1] https://fee.org/articles/new-homes-today-have-twice-the-square-feet-per-person-as-in-1973/ https://fee.org/articles/new-homes-today-have-twice-the-squa...
- ro-_-b 6y ago"OOH, when included, is typically the biggest component of the CPI. However, many coun-tries exclude OOH from their CPIs on the grounds that it is too difficult to measure. Inparticular, the harmonized index of consumer prices (HICP) in the European Union (EU) cur-rently excludes OOH." [1] OOH stands for owner-occupied housing. I am aware of the fact that the US accounts far better for housing expenses in its CPI than Eurozone countries do. [1]https://epub.wu.ac.at/7039/1/WP285.pdf https://epub.wu.ac.at/7039/1/WP285.pdf
- jerrymiller 6y ago>Their issue right now is they're tracking towards 0.44%, not 2% as they intend. This is demonstrably false, the real inflation is much higher than 2%: https://chapwoodindex.com/ https://chapwoodindex.com/