Spike in defaults

I think no apology but if I remember correctly someone did post that Bondora support had said that it was due to those B-secure deals.

For anyone not familiar with B-Secure, a borrower can pay Bondora  monthly fee for the remainder o their loan term and change the non-interest rate terms roughly as they wish, for example adding five years to a five year term. Lenders don't get any of the fee money, it seems. Bondora systems were't handling this correctly, so some of those who should have defaulted were resetting their status using B-Secure to postpone default indefinitely. Posts in the Bondora forum had people seeing around a 15% increase in their defaulted loan count when a batch of loans that should have defaulted were tracked down and moved to that state in a lump.

The issue was that borrowers can get a new schedule up to 60 months including a grace period of up to 12 months (no principal payments during this period, only interest). However, these loans didn't default even after like 7 months of no interest payments.

Bondora didn't give any notification about this issue nor about fixing it around 15th-16th December (don't remember exact date). It did increase the default rate a lot for 2014. loans by about 10% amountwise, but not sure what the effect was for earlier and later loans.

There's no real easy way to differentiate this effect from regular defaults either, since this data is not available in the dataset unfortunately.

Bondora has also changed the default definition and implemented that change probably sometime during December without notification. Now it takes 74+ days to default, instead of the 60+ it used to be.

FYI it was the 15th, my spread sheets show 57 new defaults for that day giving me a total of 817 current defaults.  The 57 defaults represent a total of  659.37 Euro capital  and 100.73 accrued interest and penalties outstanding to me.

Since official default 4 of the 57 have made very small repayments - this is a lower % than is 'normal' with my defaults.

My Bondora portfolio is practically changing to charity... For last 30 days I've received roughly same amount of interest as I've lost on additional defaulted principal for the same period. So any possible gain is negated by defaults and possibly only successful recovery can turn it into positive territory ...

If you mean the outstanding principal for defaulted loans, then that's not much of an indicator of your portfolio performance usually. The problem with comparing that amount to interest received, is that when a loan defaults 100% of the 5 year principal payments are "deducted" in your calculation, while only the few months of interest payments from proper loans are accounted for. It's very pessimistic scenario.

It's especially pessimistic since most of defaults happen quite early on in the process as well, meaning that the amount of likely interest accounted for in that calculation is very minimal in most relatively fresh portfolios (up to 2+ years even).

Essentially, if you look at it at month 12 for example, you'll have accounted for 60 months of principal payments on the defaulted side from majority of all the defaults that will happen and only 12 months of interests from the performing loans that will ever happen.

While I do enjoy a situation with my own portfolio where my XIRR shows above 20% return after deducting all overdues and defaults, it'll never happen with a fresh portfolio.

I did an anlaysis on the time when loans default on Bondora a while back. The exact months and durations have probably changed a bit nowadays, but the trend should be similar with majority defaulting during first quarter of the loan duration:
rahafoorum.ee/en/defaults-happen-bondora/

I remember reading your blog post at the time and my prior analysis led to similar results... But reality now seems a bit different (as I said my portfolio maturity is 0.4 in average). Not a surprise once we have found that Bondora was cheating with B-secure defaults and not reporting them in dataset correctly.