Fist of all, the total borrowed (regardless of what it was in the past) is a reflection of current traded volume on Bitfinex and this number grew simply because more traders joined the platform recently. This is just what it is - Bitfinex recently became the most popular exchange for BTC/USD pair.
Secondly, ALL of the borrowed by traders funds you see are fully backed primarily by THEIR OWN funds. There is simply no other technical way to borrow, but to have sufficient amount in trader's own account and the leverage at Bitfinex is a measly 2.5:1 (this is "nothing" in comparison to a 400:1 or more offered by some Forex brokerages).
What "credit bubble"? What are you talking about? Why do you care and what is your grudge?
Do you have numbers to back up the first bolded? Compare to short swaps over the same period.
Just see any charts site (bitcoinwisdom.com for instance), the bars at the bottom is volume per selected period of time. I know this is stating the obvious, but why asking the question which has the obvious answer? The volume data is right there.
Regarding the volume trends on "short swaps", they HAVE grown as well. I compared as you suggested and I do see a correlation.
Please see the 3rd party's stats site at bfxdata.com. At the top left corner select "swaps stats" for BTC and for USD, open in different browser tabs and toggle between the two. You will see that amounts of borrowed BTC since the beginning of this year peaked from about 3K BTC to about 12K BTC. In fact, the increase in borrowed BTC is bigger than for USD swaps. Usd swaps grew from about 10m to 20m (first peak) and now to 25m. Importantly, when looking at these data and comparing, you have to account for for the sentiment index and the actual BTC price data. For instance, during the last month, traders became increasingly more bullish on average, this is naturally why amount of borrowed USD swaps went up, while BTC swaps went down.
And on the second, take a look at the order book. And consider order book manipulation on top of that. Consider that Stamp liquidity is gone. In a serious decline -- a flash crash that does not recover -- how much of those longs can be sustained by the book?
I'd like to see some numbers since you seem so sure. At a glance, the situation is actually quite bad.
Don't just "glance", try to analyze and see what happens over time. The order book is not carved in stone, it is changing all the time and it is absolutely always a lot "thicker" in bids, which are closer to the current price, on both ends.
Just try to experiment for yourself: Take a "snap shot" of current order book and mark the price. See what is the volume (in order book terms) which it would take for the price to change from the current value to some other value (just mark several scenarios for the price movements in any direction). Wait for the actual price change to any of your marked values and see the real trading volume between the two points. Then, compare the real volume to your original "snap shot". You WILL see that the real volume and the volume you had per order book will be always different AND it could be higher or lower.
Another thing is that Bitfinex has "hidden orders" and those are typically quite big (there is probably no point of "hiding" anything under some 5 - 10 BTC). I know there are a lot of hidden orders out there, all the time, this is just personal experience. These make the order book a lot "thicker" than it looks.
Some numbers:
First of all, the current 2.5:1 leverage restriction, plus the 13% of trader's funds reserved in margin requirement, are more than sufficient protection against the worst flash crashes ever known to BTC market. With all of the current expansion of Bitcoin adoption, the growing interest from VCs et.c., I don't think we should expect Bitcoin price to have a complete "free fall", right?. But even then, just like for any publicly traded stock for instance, there is a trading halting, which I believe is also in place at Bitfinex.
I don't have much to say about "order book manipulation", but I've read satisfactory to me comments from Bitfinex team on this elsewhere. I believe there were some recent fixes made in this respect, but generally speaking, this is about finding the optimum between having useful features and their abuse of sorts.
About Bitstamp orders routing, I personally would indeed prefer this feature to be there (at least for bigger orders), but again this is all about big "internal" volume. Trading volume at Bitfinex is sufficiently big for them to become completely independent. After all, to say that Bitfinex is "bad" because it no longer offers Bitstamp routing feature is just wrong. We do not go to BTC-e (or to Bitstamp for that matter) and complain to them that they don't offer any routing, right? BTW, why don't we?