Dinofelis, can you explain further?:
This is a very important point. You cannot secure any thing on a block chain of which the market cap is much lower than the things you want to secure. Indeed, the cost of the proof of work (or the proof of stake for that matter) is of the order of the market cap (or lower). To attack the chain, you need, in the worst case, about to invest the whole market cap (you then redo all of the proof of work). Now, for the currency itself, that would of course be ridiculous: spending more than the total market cap of all coins, to be able to attribute yourself some. But if there are things in that chain that are worth much more than the market cap, then that might very well be worth the difficulty.
If the market cap of bitcoin is now estimated at, say, $ 4 billion, then that comes down to saying that with about $ 4 billion, you redo all the proof of work (if the mined coins were mined at the price they were worth). That means that someone able to plonk down, say, $6 billion, can redo the entire bitcoin block chain. Of course, that wouldn't be worth it. But if that chain contains a contract worth $50 billion, then that changes things: if it is worth to you $50 billion to change that contract, then plonking down $6 billion is a good deal.
So the bitcoin block chain is not more secure than about its market cap.
...because the combined value of all the assets of a municipal records dept
including salaries, building, land etc. does not even come close to
the value of the property recorded therein. The Blockchain is a (hopefully)
immutable abstraction of wealth, among other things.
The block chain being decentralized, the ONLY security it has is the cryptographic impossibility or difficulty for it to be forged or altered. That difficulty can be overcome, but at a price, and that price is exactly the financial cost of the proof of work that went in it. In other words, barring smart improvements or strong technological advance, the proof of work in the block chain that secures it cryptographically, can be overcome only by producing MORE proof of work and hence spending a LARGER financial cost than the one that is securing the current block chain.
Let me put it like this: say that the current block chain (I'm taking arbitrary numbers) is protected by a proof of work of 10^22 hashes (the integrated proof of work). If you are capable to do 2 x 10^22 hashes, then you can re-write the block chain entirely, from the genesis block onward. Those hashes come with a certain cost, and assuming that the cost of mining equals the reward of mining, and assuming that the price of a bitcoin is constant, then the total cost spent to produce these 10^22 hashes is equal to the total price of all mined bitcoins, which is nothing else but the market cap.
Now, of course, the cost of mining in reality is LOWER than the reward, which means that it did cost LESS than the total market cap to produce 10^22 hashes. Also, the bitcoin price ROSE a lot, so the early mining was *cheaper* than the current price of the then mined coins. These two arguments go in the direction of telling you that the cost of 10^22 hashes is in fact LESS than the total market cap of bitcoin.
So for a price less than the total market cap, you can ENTIRELY REDO THE BLOCK CHAIN. So the upper limit is the total market cap. If you are willing to plunk down the market cap, you can redo the block chain.
In order to obtain bitcoins, that would be totally ridiculous. But things secured in the block chain might be worth more, and then there's an incentive to do so.
A centralized recording doesn't have to have this property, because the security of it doesn't come from the price of proof of work, but rather from the trust you can put in that central authority, and the law enforcement, that will make that it cannot be modified by any individual plunking down some money on the table.
Suppose that in a municipal record is written down that association X possesses a building that's worth 10 million dollars, and the notebook in which this is written down costs only $5. In bitcoin speak, it would be sufficient to buy another notebook of $5, change the owner of that building in that notebook, and hey, it is yours !
But you can't do that with a municipal record, because the employees guarding that notebook will not allow you to replace it. If you insist, they will call the police who will come and stop you from replacing the notebook.
However, in bitcoin or in any other block chain, the notebook can be replaced by anybody. The only thing that you have to do, is pay more Proof of Work (or proof of stake). The block chain IS the $5 notebook. If you buy a new notebook, you can replace it. The ONLY reason why you don't, is that the proof of work costs more than anything that is in the notebook. The only thing that is in the notebook (the block chain) are bitcoins. The price of the book (the proof of work to make a new one) is about equal to the total price of everything that is in it. In fact it is somewhat lower.
So as long as there's not more IN the book, than the cost of a new book, there's no point in doing so.
But the day that there's something IN the book that costs more than the book itself, it is worth changing it !
In fact, a central ledger also has a price to be modified: the price of bribing all the authorities that are supposed to guard it. If you can plunk down enough money so that all the authorities guarding the municipal record allow you to change it, then you've also broken the security of the system.
The security of a block chain is in its proof of work. If you can buy more proof of work, you can corrupt it. The security of a centralized ledger is the authority guarding it. If you can buy that authority (bribe it), then you can corrupt it.