Bitcoin is a very young currency / commodity and needs to find some form of equilibrium first. This will be reached once all people who potentially want to use Bitcoin do. Until then, each influx of new users will require a price correction as the supply is fixed.
So the situation is exactly reversed wrt Bitcoin.
Two nitpicks:
1. Money is created by normal and central banks, which are somewhat different types of organizations. Central banks can directly "print" money while banks can buy IOU's for future money (loans).
2. Now the difference that is created with the loans is indeed payed back and increases the money supply. It's not destroyed.
Here is a chart of money supply of USD [1]. As you can see, the amount of money in existence is increasing continuously. It's one of the core elements that influence inflation.
[1]: http://en.wikipedia.org/wiki/Money_supply#United_States
MZM from St. Louis FRED: research.stlouisfed.org/fred2/graph/?chart_type=line&s[1][id]=MZMNS&log_scales=Left
When a loan is written off, the loan is destroyed without money being destroyed at the same time. That is, writing off a loan leads to what is effectively (in hindsight) net money creation.
That actually makes a lot of sense when you think about it: Writing off a loan means that the bank gives out money without the money being paid back. Of course there's going to be a net plus of money in circulation afterwards, because the "paying back" part is missing.
When a bank creates a loan, it adds an asset and a liability to its balance sheet. The asset is the loan (that is, the bank's customer's promise to pay back the loan), while the liability is newly created money in the customer's bank account.
The money supply is defined as the sum of all money in existence, and this includes money in bank accounts. That is, when a loan is made, the money supply grows.
When the loan is paid back, the reverse happens and the money supply shrinks again.
And yes, the amount of money in existence is increasing continuously, but so is the amount of outstanding bank loans. In other words, the evidence you present is perfectly compatible with what I'm saying.
Even on the frontpage of http://bitcoin.org/en/ it's defined as an "innovative payment network and a new kind of money" and the word currency doesn't appear anywhere.
Do you also consider a cellphone to be a "phone" in the traditional sense. Could phones browse the web, edit documents, take photos, check email, etc. 20 years ago?
'money' is the more complicated concept. If bitcoin is 'money', then whether it's 'currency' or not mostly just depends, I think, on whether you want the definition of currency to include digital certificates or only physical objects. Not a very interesting question, just a matter of definitions. Now, what 'money' is, that's an interesting question.
Uh-oh don't get us started on that one, that's a can of worms!
Used to be, way back in the past, "money" was "coin or bullion in your bank's vaults" and banks' receipts for "money" starting circulating -- hence, we have the separate word currency.
Nowadays, I suppose the distinction is largely lost, other than "currency" now implicating different jurisdictions -- EUR and USD are considered different "currencies" but both are considered "money" by and large.
Money is spent or lent or hoarded, currency circulates. One necessitates the other.
Whether "hard" or "soft", in all history and today both are certainly always "credit" -- a (rightly or wrongly) trusted claim to some future consumption of some future production. Whether the "money" is paper or metal or digital. If we swap a banana for an apple, a trade is completed and settled, no money was needed. But since most of apple sellers are not also banana buyers "right here and now", this purely mental value-association, regardless the medium it's recorded on, is needed and used and traded in-place with ease and the world mostly works! Money hence is always a debt for future trade settlement -- completion of exchange.. (Whether banks should create new money for all their lending is another question, whether all savings of people should be in ambiguous claims on future production is another question. Doesn't change this most fundamental nature of "money".)
Until they aren't