Prosper Adjusts Interest Rates on Repeat Borrowers

Regular readers of this blog will know how much I like the repeat borrower segment of p2p lending. Only officially available on Prosper, this has been one of the best performing segments of borrowers in all of p2p lending, and it happens to be where I focus most of my Prosper investments.

To be clear, when I talk about previous borrowers I am referring to those borrowers who have successfully been issued with a loan on Prosper, are in good standing on that loan, and are now looking to apply for a new loan. Their previous loan may or not have been fully paid off. Prosper allows borrowers to create a second loan listing within 6-12 months as long as certain conditions are met.

Lower Interest Rates for Borrowers, Higher Returns For P2P Investors

On Friday of last week, in this detailed blog post, Prosper announced they are adjusting the interest rates on repeat borrowers. At the same time they are making adjustments to their proprietary Prosper ratings system that will mean many repeat borrowers will jump a loan grade, say from Grade D to Grade C. This will mean lower interest rates for these borrowers.

At the same time Prosper has recalculated their all-important expected loss rates so in effect these interest rate changes will result in a higher expected return for investors. For example, D-rated repeat loans will now receive an expected 13.63% return going forward versus a 12.43% before. But D-grade borrowers will now be offered a 25.74% interest rate versus 30.99% rate before (because these borrowers were an E-grade before).

If this works out the way Prosper expects this is great news for investors and repeat borrowers. My biggest complaint about repeat borrowers is that there are not enough of them. With my strict filtering criteria I only find 4-6 loans a week to invest in and I would like to see many more loans. Hopefully, this change will increase the number of available loans that meet my criteria while maintaining the high ROI on these loans.

No More Repeat Borrowers with HR Grade

The interesting part is that interest rates have not dropped across the board. In fact, some interest rates for previous borrowers have increased slightly. But with the changing calculation for assigning a loan grade it will mean reduced rates for most repeat borrowers.

Also, along with this change repeat borrowers with an HR rating are no longer allowed. The reason for this seems to be that eligible borrowers that would have had an HR rating will now be classified with an E rating.

Why Previous Borrowers on Prosper Are So Good

You just need to look at the data on Lendstats to appreciate previous borrower loans. Looking at Prosper 2.0 loans (those originated from July 2009 onwards) Lendstats has an estimated ROI on all loans (as of today) at 9.33%. When looking at all previous borrower loans that estimated ROI jumps to 13.14%, almost four full percentage points.

When you look at the risker loan grades this difference is even greater. All D,E and HR loan grades on Prosper 2.0 are returning an estimated ROI of 11.93% at Lendstats. When looking at just previous borrower loans in those grades the estimated ROI jumps to 16.95% or more than five percentage points higher. Clearly this is a very good performing segment and one that has probably been overpriced for some time.

My Own Experience with Previous Borrowers

I started investing in previous borrower loans almost exclusively on Prosper about six months ago. My average age of these loans in my portfolio is just 4.5 months so I know these loans are still very young. But the thing that has amazed me about my previous borrower loans is that they are all still current. There is not one late loan among the 81 loans that I have invested in. And I am investing in only the risker loans – those loans rated D, E and HR with an interest rate of more than 20%.

I have invested in these loans expecting a default rate of more than 10%. Of course, maybe I will end up seeing defaults close to this number but when I look at my own portfolio on Prosper and see every single loan current I am very pleasantly surprised to say the least.

These changes in borrower ratings and interest rates are quite complicated and I don’t really have the space here to cover it fully. I highly recommend that investors read the PDF file Prosper released that contains a thorough explanation of these changes.

It will be very interesting to see how these changes play out a few months and years from now. What do other Prosper investors think? Is this a good move by Prosper? I am always interested to hear your thoughts.

 

Subscribe
Notify of
29 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Ryan
Ryan
Oct. 10, 2011 4:18 pm

My intuition was that these loans were going to exceed Prosper’s profit expectations. While I’ll be sad to see this extra profitable segment of the business have its rates decreased to a more standard profit level, it should encourage responsible borrowers to return to Prosper in the future.

Dan B
Dan B
Oct. 10, 2011 5:15 pm

Make no mistake, this move is 100% for the borrower. The rest is just smoke & mirrors & spin. It makes a lot of sense to do this since loan originations are where 90%+ of the money is made in p2p. Assuming the Prosper average 36 month term, $6k loan………….the average repeat borrower saves roughly $15-16 from their monthly payment.

The smoke & mirror & spin part is to suggest that the investor also benefits. Loss rates are what they are & will be what they will be. Loss rates will not get better or worse just because someone assigns a different letter grade to them. If loss rates were in fact better than previously estimated then investors were already reaping the benefits of that fact regardless of any “estimates” & certainly before any of the changes made here…………….. & most importantly at the higher interest rate. To suggest that investors will now benefit additionally going forward by a lowering of those interest rates is nonsense imo. How can the investor benefit by getting less interest? Keep in mind that none of the realities of life will have changed in any way for any of these borrowers except they will now pay $15-16 less on average per month. Will that decrease make an observable decrease in the loss rates of these new loans? Prosper will have you believe that it will & that it will in fact not only negate the roughly 5% drop in the interest rate investors receive on these loans but more than make up for it…………….thereby giving you a better actual return. Come on, really??

I have no idea why Prosper has decided to spin it in this fashion, in effect trying to have it both ways. What’s wrong with just being straight? Just say here is this group of repeat borrowers who have proven with their repayment history that they are a better credit risk than we had initially estimated. Therefore in future we’re raising their credit grade & lowering the interest rates they have to pay. What’s wrong with that? If all the above was correct & true, investors could/would accept that, wouldn’t they? I think they would. So why throw in the……………we’re lowering these rates but you’ll get a higher return BS?

Lou lamoureux
Lou lamoureux
Oct. 10, 2011 5:29 pm

@DanB it’s marketing! Why say something is better when you can say it is new and improved with proprietary doodads that whiten teeth, give shinier hair, AND better returns.

Did they really need to announce the change? No. Making it sound earth shattering in a blog post/ press release might get a media outlet to pick it up.
Lou

Charlie h
Charlie h
Oct. 10, 2011 6:16 pm

What was I saying about misspriced loans on prosper again? 🙂

This change should help prospers loan origination. It does not change the ACTUAL credit risk of the borrower.

Roy S
Roy S
Oct. 10, 2011 7:16 pm

, I’m glad your strategy is working out for you. I, on the other hand, only have 3 loans currently late. 100% of them are repeat borrowers! Luck of the draw or possibly your strategy of picking repeat borrowers is just superior to mine? I’m hoping the former, but I fear it may be the latter.

@Everyone else, I think the main point is that Prosper has adjusted their risk model to better price loans based on a borrowers profile. If they are correct then they can offer lower rates to certain groups of people because the loss expectations for that group is lower. Basically what this means is the can better price risk. So if their previous model had 50% of the people rated D who actually deserve a D grade loan but also had 30% who should have been rated E and 20% who should have been rated C, removing the 50% who are not D’s would mean that someone who has a higher chance of defaulting would be correctly rated an E and someone who has a lower chance of defaulting would be correctly rated a C. Now you have someone who should be paying a higher interest rate paying a higher interest rate. Would you rather be investing in a D note with an E borrower or a D note with a D borrower? Conversely, you also have someone who should be paying a lower interest rate paying a lower interest. If you look back at when Peter took out a loan, he received an E grade with an E interest rate. Obviously, he could easily get a loan elsewhere for a far lower interest rate and under normal circumstances would not have taken out the loan. Assuming that Prosper now correctly prices his loan, he may be more likely to take out a loan from Prosper than he would have been prior to the changes (i.e. an B borrower would not take out a D grade note since the B borrower could go elsewhere for a lower rate). Of course, this again assumes that their models are getting better and they can price risk more accurately. Like so many other conclusions we seem to come to on this board, we’ll find out when the notes come due.

Dan B
Dan B
Oct. 10, 2011 7:17 pm

Lou Lamoureux……..Yeah, I know, but it would still be refreshing to hear it without the marketing once in a while. In a perfect world, right?

Charlie H………..Exactly.

Dan B
Dan B
Oct. 10, 2011 7:20 pm

Roy S……….Granted, but in no way does any of this translate to investors getting a better return due to this change. And that’s the only part I’m arguing.

Bilgefisher
Oct. 11, 2011 7:46 am

On one side, I think it encourages more repeat borrowers. That is good.

That said, I agree 100% with Dan. It is pure marketing. I see no benefit to investors that invest outside the grades. Lower interest rates for the same risk. How does that benefit investors exactly?

Jason

Roy S
Roy S
Oct. 11, 2011 10:42 am

@Dan, I skimmed the article on Saturday morning, and I didn’t refer back to it. The article does seem to focus on lowering interest rates for repeat borrowers. I made the leap that they were updating all their risk models resulting in more accurate pricing across all loans. If they are simply updating it for repeat borrowers by rewarding the repeat borrowers with better rates, then you are correct. The default curve for that subset remains constant while the interest rate is lowered resulting in lower returns for lenders. It does appears as though they are trying to obfuscate the fact that the risk associated with this subset remains constant while the rate is lowered (i.e. the Annual Expected Loss Rate remains constant for this subgroup of loans but they are stating it changes probably based on the rates for the entire new grade to which they are being assigned and not for this specific subset). I’m looking at their pdf, but I would need more time to analyze it (which I do not have at this moment), but it that looks like they recognize the fact that moving these better performing loans to a better grade means that it leaves the remaining loans with the old grade having a higher loss rate and that they are offsetting that with higher pricing (page 7, https://www.prosper.com/prm/pdf/ProsperLossRateandPricingChanges-October2011.pdf). I just don’t have time at the moment to go over it.

On a final note, I do agree with their statement: “We expect that better rates for repeat borrowers will attract more qualified borrowers to Prosper, so that our investors will enjoy more and higher quality loans.” But that can be said of more accurate UW models in general and not just to repeat borrowers.

Charlie H
Charlie H
Oct. 11, 2011 11:01 am

Making a more accurate Annual Expected Loss Rate does not change the actual lose rate.

If they lower the AELR by an amount larger then the drop in the interest rate then they can “show” an estimated higher return.

I think this is the crux of what they are “spinning”.

Dan B
Dan B
Oct. 11, 2011 2:56 pm

Peter…………..Whether Prosper “estimates” a higher or lower ROI in no way affects the actual ROI. That’s not really a debatable point simply because it’s an “estimate”. Whether someone erroneously mistakes the “estimate” as being “fact” doesn’t affect the actual ROI that will occur either, though I suppose you could make the case that a casual investor may be persuaded to invest due to a higher estimate.

But the point that I & some others here are making is that there’s just no way to reasonably suggest that one can take a group of borrowers, lower the interest rate they pay to lenders & then suggest that the lenders are somehow going to make more money because of these changes. It just makes no sense. If anything else one can argue (& someone else already has) that these changes will in reality bring down the ROI within this group so that it’s more in line with the average ROI.

Dan B
Dan B
Oct. 12, 2011 8:19 am

Peter……….Perhaps I’m way off my game right now, but I’m having a real tough time following the path of your argument.

Roy S
Roy S
Oct. 12, 2011 10:38 am

Peter, I am trying to find that 13.63% return. What I am seeing is that the expected return per Prosper comment image) is actually decreasing from from 12.43% to 12.06%, which they don’t draw attention to. Further, they are attempting to draw our attention to the Return:Risk change from 1.24 to 1.42 stating that the return is higher compared to the risk we are taking. The problem here is that they are using the original expected loss rate of 10.00% to calculate the original 1.24 Return:Risk ration. Looking at page 6 on https://www.prosper.com/prm/pdf/ProsperLossRateandPricingChanges-October2011.pdf they specifically state, “Before, the expectation was a 10% loss rate for a repeat loan from the same borrower, which would assign them to a D rating. Experience now brings expectations down to an 8.5% loss rate, which puts them into the C rating.” This specifically states that the original expected loss rate of 10.00% was incorrect and the newly revised risk rate of 8.5% is the correct loss rate for the original loans. Instead of using that loss rate (which Prosper knows to be true per their pdf file) they chose to remain with the original “expected” loss rate rather than the real (i.e. actual loss rate) of 8.5%. When you use the actual loss rate instead of the 10.00% expected loss rate, the Return:Risk for the “Second Loan (OLD)” is actually 1.46 NOT 1.24. So the Return:Risk ratio actually decreases from 1.46 to 1.42 (i.e. investors are receiving a lower return for the same risk).

Sorry for the long round about way of saying that the risks have not changed, but the return (interest rate) has decreased, but I needed to use that longer explanation of why people on here are upset with the way Prosper is presenting their change of interest as beneficial to investors. For me, the more I look at this the more upset I am with Prosper for trying to put a positive spin on this change for investors. Really, a change from 1.46 to 1.42 isn’t that much worse, but Prosper attempting to hide that and then put a positive spin on it does upset me and sours me a little on them, too.

DutchInChicago
DutchInChicago
Oct. 12, 2011 10:56 am

One of the drawbacks as an investor is that a current (none repeat borrower) is getting an offer to refinance at a lower rate as a repeat borrower. So I loose my high interest loan and have it replaced with a lower interest loan but I have the same risk exposure.

Still it was obvious that repeat borrowers were miss priced before and the previous situation was too good to be true.

Roy S
Roy S
Oct. 12, 2011 10:56 am

Actually, as an addendum, if the expected loss rate was 8.5% rather than 10.00% then the actual return is higher than the 12.43% expected return. This in turn increases the Risk:Return ratio as well, making the 1.46 a low figure for what the actual Risk:Return number should be and upsets me further…if anyone else wants to calculate it to make me further upset go ahead. I might do so later, but not at this time.

Roy S
Roy S
Oct. 12, 2011 1:55 pm

Ah, I see where you got the 13.63% return. It looks like you pulled the wrong number. A loan that was under the old calculations a D loan is now a C loan under the new matrix, so the correct expected return is 3 lines up at 12.06%.

I am still having trouble finding the actual returns for this subset of loans. The best I can find is a 15.7% actual return for all D notes per Prosper’s website for “Loans Originated July 2009 – August 2010.” Since this post states that the repeat borrowers are performing better than their peers, I will assume that 15.7% is a low figure for the average actual returns for repeat D loan borrowers. Now using the 8.5% loss rate for this subset of loans, I calculate a 1.847 Return:Risk ratio. With Prosper’s new Return:Risk ratio (given the future expected returns and future expected loss rate) of 1.42, it appears as though lenders are significantly worse off (not better off) than Prosper spins in their blog.

Again, none of this would be an issue with me if Prosper were more upfront about this than having all of us do our DD to find out that the post is, in my opinion, trying to spin this change. I think loans should be competitively and accurately priced. If Prosper were to simply have come out stating that they are overpricing the loans compared to the actual defaults coming in and they are updating the UW models and more accurately pricing the notes, it would not be such a big deal to me. Of course, those of us on the board are probably a minority to those who are actually going through the trouble to figure all this out.

Dan B
Dan B
Oct. 12, 2011 2:06 pm

Peter……..Thank you, I understand now. It’s kind of subtle, almost smart & even a bit insidious, all at the same time. I like it!

Roy S
Roy S
Oct. 12, 2011 3:08 pm

I think I will have to still disagree with you, Peter. Again, using the actual numbers (or as close to them as I can get), a person who invested in any D loan prior to the change would have a return of 15.7% (again, all the information points to it being higher for repeat D loan notes, but I am using this number as a conservative number) and an actual loss rate of 8% giving them a 1.847 Return:Risk ratio. If a person were to invest in the same type of D note (one which would have been previously scored as an E note under the old matrix but a D note under the new matrix), the expected return is the 13.63% as you noted, but the expected loss rate is 10.50% (page 4 of the previously linked to pdf file). This gives a Return:Risk ratio of 1.30 which is even lower than 1.42 if you were to invest in a repeater C note that would have been scored as a D note under the old matrix. Again, I don’t have a big problem with the actual change just with the marketing/presentation.

Roy S
Roy S
Oct. 12, 2011 3:15 pm

This doesn’t change the 1.847 Return:Risk ratio, but in my most recent comment I stated 8% instead of the correct 8.5% which my other posts are correctly showing.

Also, just looking at the actual return of 15.7% for D notes compared to the expected to the expected return of D notes under the new matrix of 13.63%, I still find it difficult to state that the average investor in D note repeater loans under the new matrix will be performing better than those who invested in D note repeater loans under the old matrix, let alone any D note regardless of previous loan status.

Dan B
Dan B
Oct. 12, 2011 7:25 pm

Peter………..I would have thought that as an advocate of p2p investing, you would be opposed to any change where you believe that “overall investors will be worse off”. Furthermore I’d have thought that you would NOT be comfortable with any type of spin that suggests that the average investor will benefit from these changes, when as you’ve stated, they will not.

Dan B
Dan B
Oct. 13, 2011 8:09 am

You know I’m not even sure that Prosper will get many more repeat customers from this move. Think about it…………a 5 year old company that deals in 3 year average loans where 34% of new loans issued this year are to repeat borrowers already. That’s a pretty high percentage number for a young loan company. I mean it’s not like they’re selling groceries here. How many more repeaters do they think they can get? Anyways……………………

Roy S
Roy S
Oct. 13, 2011 9:14 am

, Yeah, their marketing makes it look like it’s a good thing for the investor because the expected return is going up. If there were no historical data for either population, then you can say that returns look like they will be better for investors under the new pricing matrix. But when you take a look at the actual returns that investors have been receiving and compare that to the expected returns they will be receiving under the new matrix, investors returns will actually be decreasing for this subset of loans. And that is where I have a beef with Prosper.

Where we do agree, Peter, (and what I have attempted to make abundantly clear) is that I believe more accurate and competitive pricing of loans will benefit everyone–Prosper, lenders and borrowers–even when the returns for a specific subset of the loan population decreases. When I first skimmed the blog post on Saturday morning, that is what I initially thought the post was about. It wasn’t until others on your blog brought it to my attention that the post was attempting to (in my opinion) spin this change and mislead investors into believing Prosper’s new matrix for repeat borrowers would benefit investors directly by increasing returns for this specific subset that I really took a look into. So much for me giving Prosper the benefit of the doubt when it comes to their marketing.

Dan B
Dan B
Oct. 13, 2011 11:09 am

Look I think I speak for everybody here in saying that we all want Prosper to succeed. Even those of us who may not be investors there wouldn’t enjoy the fallout that would likely occur if they don’t succeed.

I just think that Prosper doesn’t do itself any favors being so creative with their facts when they don’t need to be. Situations like this just further reinforce the feeling that every single word, every single number needs to be fact checked & put through a strong spin filter, & that is the main reason why I’ve never recommended any investor to Prosper.

What is perplexing is that returns in the 8-9% range are almost universally characterized as being very good to excellent in almost all market environments, & that is what Prosper appears to be producing these days. So that’s plenty, you don’t need to spin or exaggerate if you can produce those types of numbers long term.