Bondora review after 6 months
Jun 21, 2017 9:50:03 GMT
JamesFrance, coolrunning, and 6 more like this
Post by kulerucket on Jun 21, 2017 9:50:03 GMT
I thought that I would share my experiences of Bondora of the last 6 months. As you may have read from the earlier investors, it is the general opinion that people are lucky to get their money back and after exiting they are left with a large collection of defaulted loans only trickling money from recoveries. While I have taken such reviews on board, I could not find any real information from newer investors and a lot has changed in Bondora since a few years ago (new loan ratings, new collection rules etc). I wanted to invest a test amount for enough time to gather information on whether Bondora is now a better platform or still to be avoided.
So first of all the background information. I invested 500€ in December 2016, I enabled the Portfolio Manager and opted for the "Balanced" risk-return strategy. After taking a few weeks to fill up with 5€ loans, I ended up with the following loan distribution (by value):
Initially the HR loans filled up first due to the fact that there are more of them available that other loans so the initial distribution was scary. Initially, when some HR loans started to get more overdue, I started to think about the effect that defaults would have over time. Would the PM keep buying an even distribution of loans as the higher risk failed to repay, or would it take the unpaid principal into account? The former would cause an ever increasing number of HR loans so for this reason, I tracked the distrbiution over time:
As you can see, there is some jostling, but overall no increase in the proportion of HR loans.
My plan was to let this run for 1 year, reinvesting the interest earned and letting the PM do it's thing. I figured that after a year I should be able to see enough of a trend to figure out what was going on. However after 6 months I think I have a pretty good idea already.
So what is the status today?
According to Bondora, my net return is reported as 26.89%, and this puts me in the top 7.8% of investors. In fact my statistics page looks like the returns are way over what the predictions are for the PM and steadily increasing.
However my own picture looks quite different. In Bondora world, a defaulted loan only counts the principal missed so far in terms of money lost. For HR loans of say 60 months at 150% interest, the replayment for first year are nearly 100% interest, meaning that even though a loan has defaulted, it counts as virtually no loss. The hit only comes after 3-4 years when the principal payments missed start ramping up. So IMO, you need to look much more closely at defaults.
Over time, the amount of delayed/defaulted loans has steadily increased. See below:
As of today my loan status distribution looks like this:
In real numbers then, 64.10€ is in default and a further 19.92€ is in the Cure Period (between 61 and 74 days late) which mean defaulted but not yet red. My account value at this time is 561.77€. Due to the way that Bondora deducts overdue principal from the account value, the interest earned is roughly the same as the amount in default. In the near future I will have more defaulting loans and the interest will not be enough to match them.
Now, loans that will default are more likely to default early and the longer that payments are made, the less likely it is that a loan will default. This means that my interest should catch up with the defaults again and I already see the rate of default slowing. By my estimation, the interest will match the defaults outstanding by the end of the year (assuming no recovery).
It is therefore clear, that as longer term investors have been saying all along, the profits to be earned are completely driven by the recovery of defaulted loans. So what about my recovery? Well according to my recovery statistics, the recovery done by Bondora is astounding at first glance:
Yes that's right, 2 thousand percent recovery in May! What does this really mean though? Well it means that out of about 12 defaulted loans, one was recovered in full (both principal and interest) and paid 8.65€, whereas the principal missed from the defaults (remember that most payments are nearly all covering interest) was 0.43€ [8.65€/0.43€*100 = 2011%]. Now, I grant that a full recovery is a good result for this loan but if someone asked my how much of your defaults recovered? I'd answer about 1/12 (or 8%). When I see these recovery rate stats sent out by Bondora and see 50-70% recovery rate, I assume that overall this just means that a very low number of loans must have been recovered.
In the end, I do believe that more of these loans will recover in time. Approx 30% are home owners and some have court orders etc on them. It will be slow though. In addition, if generate a future cashflows report, assume all defaulted loans will be pay 0€ and all non-defaulted will pay on schedule, I still end up with an XIRR of ~10% [*see EDIT at end for more detail]. However as things stand, this could take a very long time and tie my money up for years. Any new loans bought now will eventually follow the same pattern and the fact that the first year of any paying loans only really cover the defaulting loans bought in the same period and average out to a 0% return, only to generate more profit on average in the later stages after the defaults from the same initial period slows down.
For this reason, I've now decided I've seen enough and have disabled my PM to start a slow withdrawl, letting things wind down on their own. I have no need to sell everything off and lose all of the current loans that are on track. I consider my experiment worth the effort and I should at least get my money back eventually. But there are other platforms offering lower default rates, faster returns and cheaper/easier exit strategies out there.
EDIT:
[*]: When analysing the XIRR on future payments of different loans, I have calculated the following:
XIRR where "current" loans pay to end, all others written off: 12.6%
XIRR where non-default (<=60 days late) pay to end, all others written off: 26.2%
In the long term this is still quite a positive result. However since I have only just stopped the Portfolio Manager, the newest loans have not yet contributed to the defaults. This is why in the text above I have estimated 10%. A more accurate figure will be available once the PM has been off for a few months.
So first of all the background information. I invested 500€ in December 2016, I enabled the Portfolio Manager and opted for the "Balanced" risk-return strategy. After taking a few weeks to fill up with 5€ loans, I ended up with the following loan distribution (by value):
| A | B | C | D | E | F | HR |
| 6.9% | 16.8% | 22.8% | 17.8% | 15.8% | 5.9% | 13.9% |
Initially the HR loans filled up first due to the fact that there are more of them available that other loans so the initial distribution was scary. Initially, when some HR loans started to get more overdue, I started to think about the effect that defaults would have over time. Would the PM keep buying an even distribution of loans as the higher risk failed to repay, or would it take the unpaid principal into account? The former would cause an ever increasing number of HR loans so for this reason, I tracked the distrbiution over time:
As you can see, there is some jostling, but overall no increase in the proportion of HR loans.
My plan was to let this run for 1 year, reinvesting the interest earned and letting the PM do it's thing. I figured that after a year I should be able to see enough of a trend to figure out what was going on. However after 6 months I think I have a pretty good idea already.
So what is the status today?
According to Bondora, my net return is reported as 26.89%, and this puts me in the top 7.8% of investors. In fact my statistics page looks like the returns are way over what the predictions are for the PM and steadily increasing.
However my own picture looks quite different. In Bondora world, a defaulted loan only counts the principal missed so far in terms of money lost. For HR loans of say 60 months at 150% interest, the replayment for first year are nearly 100% interest, meaning that even though a loan has defaulted, it counts as virtually no loss. The hit only comes after 3-4 years when the principal payments missed start ramping up. So IMO, you need to look much more closely at defaults.
Over time, the amount of delayed/defaulted loans has steadily increased. See below:
As of today my loan status distribution looks like this:
| Current | 1-7 | 8-15 | 16-30 | 31-60 | 61-90 | 91-120 | 121-150 | 151-180 | 181+ |
| 74.33% | 6.95% | 3.40% | 1.23% | 0.89% | 4.43% | 2.64% | 6.12% | 0.00% | 0.00% |
In real numbers then, 64.10€ is in default and a further 19.92€ is in the Cure Period (between 61 and 74 days late) which mean defaulted but not yet red. My account value at this time is 561.77€. Due to the way that Bondora deducts overdue principal from the account value, the interest earned is roughly the same as the amount in default. In the near future I will have more defaulting loans and the interest will not be enough to match them.
Now, loans that will default are more likely to default early and the longer that payments are made, the less likely it is that a loan will default. This means that my interest should catch up with the defaults again and I already see the rate of default slowing. By my estimation, the interest will match the defaults outstanding by the end of the year (assuming no recovery).
It is therefore clear, that as longer term investors have been saying all along, the profits to be earned are completely driven by the recovery of defaulted loans. So what about my recovery? Well according to my recovery statistics, the recovery done by Bondora is astounding at first glance:
Yes that's right, 2 thousand percent recovery in May! What does this really mean though? Well it means that out of about 12 defaulted loans, one was recovered in full (both principal and interest) and paid 8.65€, whereas the principal missed from the defaults (remember that most payments are nearly all covering interest) was 0.43€ [8.65€/0.43€*100 = 2011%]. Now, I grant that a full recovery is a good result for this loan but if someone asked my how much of your defaults recovered? I'd answer about 1/12 (or 8%). When I see these recovery rate stats sent out by Bondora and see 50-70% recovery rate, I assume that overall this just means that a very low number of loans must have been recovered.
In the end, I do believe that more of these loans will recover in time. Approx 30% are home owners and some have court orders etc on them. It will be slow though. In addition, if generate a future cashflows report, assume all defaulted loans will be pay 0€ and all non-defaulted will pay on schedule, I still end up with an XIRR of ~10% [*see EDIT at end for more detail]. However as things stand, this could take a very long time and tie my money up for years. Any new loans bought now will eventually follow the same pattern and the fact that the first year of any paying loans only really cover the defaulting loans bought in the same period and average out to a 0% return, only to generate more profit on average in the later stages after the defaults from the same initial period slows down.
For this reason, I've now decided I've seen enough and have disabled my PM to start a slow withdrawl, letting things wind down on their own. I have no need to sell everything off and lose all of the current loans that are on track. I consider my experiment worth the effort and I should at least get my money back eventually. But there are other platforms offering lower default rates, faster returns and cheaper/easier exit strategies out there.
EDIT:
[*]: When analysing the XIRR on future payments of different loans, I have calculated the following:
XIRR where "current" loans pay to end, all others written off: 12.6%
XIRR where non-default (<=60 days late) pay to end, all others written off: 26.2%
In the long term this is still quite a positive result. However since I have only just stopped the Portfolio Manager, the newest loans have not yet contributed to the defaults. This is why in the text above I have estimated 10%. A more accurate figure will be available once the PM has been off for a few months.
loans.