### Author Topic: evaluating ytm  (Read 5274 times)

#### mikedev10

• Newbie
• Posts: 37
##### evaluating ytm
« on: January 02, 2019, 12:55:43 PM »
hey fellas, could use some help here;

first off i am curious as to why any notes are even listed for negative ytm or near 0 ytm, why would anyone purchase these at all?

my real challenge is in deciding what amount of ytm is "worth it" - there are A grade loans at 3, 4, 5% YTM, other values at B, and i'm having a hard time coming up with "draw the line here" which i assume is actually "draw the line here based on the default level of risk."

so rather than "buy A if ytm is 4.5 or above"

i'm thinking my rule should be "do something with original interest rate and months remaining, and if this is better or equal to ytm, and tm is 4.5 above, then buy"

am i overthinking this?  or is my rule for "this is still pricing default risk effectively" going to be much more arbitrary than a more standard calculation i'm expecting?

#### Rob L

• Hero Member
• Posts: 2100
##### Re: evaluating ytm
« Reply #1 on: January 03, 2019, 02:05:05 PM »

#### mikedev10

• Newbie
• Posts: 37
##### Re: evaluating ytm
« Reply #2 on: January 03, 2019, 03:26:11 PM »
thanks.  seems like i have been overthinking it a bit and should just assume original risk is priced in and i should feel good purchasing a note at no markup or a discount, particularly if the credit rating has stayed neutral or gone up.

#### Fred93

• Hero Member
• Posts: 2203
##### Re: evaluating ytm
« Reply #3 on: January 04, 2019, 01:48:33 AM »
thanks.  seems like i have been overthinking it a bit and should just assume original risk is priced in and i should feel good purchasing a note at no markup or a discount, particularly if the credit rating has stayed neutral or gone up.

There's an effect that you need to understand, which Rob L was trying to explain with those curves.

As the remaining fraction of the loan (which is what you are buying) gets shorter, you need more and more "discount" to get the same return as if you had bought the original note.  The reason for this is that LC takes a fee of 1% of each note payment.  Another way to think of this is that it is 1% of all the money you're going to get from the borrower.  Part of each payment is principal simply being returned to you, and part of each payment is interest.  So strangely, part of LC's fee is 1% of your own money being returned to you.  If loans are all full term, ie say 36 months, then this always works out the same, because at a given interest rate and term, the ratio between total interest and principal paid over the life of the loan is a constant from loan to loan.  However, when you buy a short "tail" of a loan in the secondary market, something different happens.  The fraction of payments which is principal is higher, and the fraction which is interest is lower.  That means that the fee you pay is a higher fraction of the interest you receive (which is of course the source of your return).  Therefore the NET income (interest minus fees) is lower.

To counter this effect, you need a steeper discount on notes with a smaller number of payments remaining.

This is true at all risk levels.  This is why LC calculates a YTM.  Theoretically, you can see this effect by looking at YTM.  Notes with smaller numbers of payments remaining require a larger discount to get a reasonable YTM.

However, none of us have ever been able to duplicate LC's YTM calculation.  Therefore, IMHO, I think it is wrong, and I calculate my own.  Your mileage may vary.

As for your original question of why notes are listed with negative YTM ... The answer is simple.  People are stupid.  People list notes at all sorts of crazy prices.  Most people can't do the math, so they list notes at a price they figure is high to see if any sucker's buy, and then over time they lower the price to see what happens then, rinse, repeat.  ALMOST ALL NOTES LISTED are listed at stupidly high prices, because if they were listed at reasonable prices, I would have bought them already.  What looks to the naive user as a big inventory is actually a big pile of irrationally priced notes that are not really indicative of any sort of market pricing.

The rule you suggested in your original message is appropriate.  That's what I do.

#### Roux

• Newbie
• Posts: 24
##### Re: evaluating ytm
« Reply #4 on: January 04, 2019, 12:37:23 PM »
It is very difficult to match LC’s YTM calculation exactly, however we have come extremely close. We have an in depth article on our blog explaining the formulas and process.

https://blog.liquidp2p.com/a-fair-trade-how-liquid-p2p-prices-lending-club-notes-for-the-secondary-market/

Within the article is another link to an amoritization schedule detailing LC’s fees and their effect on YTM written by our Data Scientist.

https://www.liquidp2p.com/yield-calculation
« Last Edit: January 04, 2019, 12:41:48 PM by Roux »

#### Rob L

• Hero Member
• Posts: 2100
##### Re: evaluating ytm
« Reply #5 on: January 04, 2019, 02:26:29 PM »
It is very difficult to match LC’s YTM calculation exactly, however we have come extremely close. We have an in depth article on our blog explaining the formulas and process.

https://blog.liquidp2p.com/a-fair-trade-how-liquid-p2p-prices-lending-club-notes-for-the-secondary-market/

Within the article is another link to an amoritization schedule detailing LC’s fees and their effect on YTM written by our Data Scientist.

https://www.liquidp2p.com/yield-calculation

As good luck would have it the A1 interest rate used in the original post first response (OPFR) is the same as the interest rate used in Table 1 in your yield-calculation paper (5.32%, 36 month term). A quick glance shows that the declining YTM's caused by LC's fees are present in both the OPFR and your Table 1, but the YTM's calculated are quite different. For example for month 1 (no payments yet made) the YTM of the OPFR is 4.65%. The month 1 YTM_w_fee in Table 1 is 4.43%. Quite a large difference. Why?

In general all the YTM's of the OPFR are higher than the YTM_w_fee's of Table 1. It appears that the Table 1 column "fee", which is 0.01 for all months, is used in the calculation as the amount (\$0.01) that LC charges rather than a percent of payment (1.00%). The column my_interest is the amount of amortized interest for each month minus \$0.01. The OPFR used the amortized interest for each month minus 1% of the payment (ie 1% x \$0.743 = \$0.00743)? IIRC LC has a minimum fee of \$0.01. If so then Table 1 shows the correct values, otherwise the OPFR values are correct. A good example of how small things can cause different results in YTM calculations.
« Last Edit: January 04, 2019, 02:29:07 PM by Rob L »

#### Roux

• Newbie
• Posts: 24
##### Re: evaluating ytm
« Reply #6 on: January 04, 2019, 09:31:29 PM »

Quite a large difference. Why?

As you have already noted, our example table is based on a \$25 note and thus subject to the rounding of its service fees to the nearest whole cent.

Lending Club Help

Quote
All fees are calculated to the tenth decimal place and will appear rounded to the nearest whole cent in your account.
https://help.lendingclub.com/hc/en-us/articles/215480768-What-fees-does-LendingClub-charge-investors-

#### Rob L

• Hero Member
• Posts: 2100
##### Re: evaluating ytm
« Reply #7 on: January 05, 2019, 12:08:36 PM »

Quite a large difference. Why?

As you have already noted, our example table is based on a \$25 note and thus subject to the rounding of its service fees to the nearest whole cent.

Lending Club Help

Quote
All fees are calculated to the tenth decimal place and will appear rounded to the nearest whole cent in your account.
https://help.lendingclub.com/hc/en-us/articles/215480768-What-fees-does-LendingClub-charge-investors-

The LC article to which you referred makes it clear that fees are computed to the tenth decimal point. All amounts and balances are computed and maintained to that same precision. I didn't see any mention of a \$0.01 minimum fee. Don't know why I thought there was. Amounts rounded to the nearest cent are for display purposes only; hence the word appear.

Quote
To see the full amount of the fees calculated to the tenth decimal place, you can hover your cursor over the dotted line below the rounded number.

For example see below:

Hovering over any of the underlined currency columns reveals a value to the tenth decimal place.
So, I would argue that the YTM_w_fee values in Table 1 are incorrect.
Wish it were not so and would like to be convinced otherwise. Wouldn't be the first time I was wrong. Ask anybody

#### Roux

• Newbie
• Posts: 24
##### Re: evaluating ytm
« Reply #8 on: January 12, 2019, 09:17:01 AM »
Rob L you are correct. We rounded the fees to the nearest whole cent, when we should have kept them to the tenth decimal place for our calculations. I did call LC Investor Services just to confirm this as well. Your spreadsheet is accurate, ours will need to be revised with lower LC investor fees. I would assume the two will match up once we correct it. Thank you for your detailed posts.