sharetrader
Page 259 of 465 FirstFirst ... 159209249255256257258259260261262263269309359 ... LastLast
Results 2,581 to 2,590 of 4649

Thread: Harmoney

  1. #2581
    Member
    Join Date
    Aug 2017
    Posts
    212

    Default

    How can F5 possibly be profitable?

    According to the forecasts, annual default rate of 9.49 annually, applied over 5 years gives a loan default rate of 47.5%. The hazard curve then suggests So 24% default against 30% interest in the 1st year.

    2nd year and onwards, a severely eroded premium base (even without early repays) , so fall short of recouping default losses through remaining cashflows.

  2. #2582
    Member
    Join Date
    May 2016
    Posts
    236

    Default

    Quote Originally Posted by leesal View Post
    How can F5 possibly be profitable?

    According to the forecasts, annual default rate of 9.49 annually, applied over 5 years gives a loan default rate of 47.5%. The hazard curve then suggests So 24% default against 30% interest in the 1st year.

    2nd year and onwards, a severely eroded premium base (even without early repays) , so fall short of recouping default losses through remaining cashflows.
    Simplified example in round numbers ignoring repayments and arrears. Say you invested $100k (fully diversified). After one year you would expect 10 to have gone bad and 90 to have paid interest at 30% ($27k) so you would be up $17k before fees and tax and arrears. The next year the same %'s apply but to a lower base so the return would be the same. Personally I think there is better value in the other grades. It's not just the expected return but also the variability that matters.

  3. #2583
    Member
    Join Date
    Aug 2017
    Posts
    212

    Default

    Quote Originally Posted by RMJH View Post
    Personally I think there is better value in the other grades. It's not just the expected return but also the variability that matters.
    Agreed

    Quote Originally Posted by RMJH View Post
    Simplified example in round numbers ignoring repayments and arrears. Say you invested $100k (fully diversified). After one year you would expect 10 to have gone bad and 90 to have paid interest at 30% ($27k) so you would be up $17k before fees and tax and arrears. The next year the same %'s apply but to a lower base so the return would be the same.
    Hazard Curve


    The forecast default rates are shown as a consistent annual rate over the term of the loan. In reality, defaults are more likely to follow what is known in statistical terminology as a time-varying hazard rate. This chart shows the profile of the hazard curve of the personal loan portfolio to date and shows that almost 60% of the defaults that have occurred may been within the first 15 months of the loan.



    If Harmoney's marketing is to be believed, 50% of the defaults occur within 12 months. At grade F5 that gives- 24%. So in the simplistic model 24 gone and 76 paying 30% interest 23%. ~ negative (1%) return.

    Year 2 resets itself- 25% more of total defaults - 12 gone, and remaining 64 paying 30% (19.2) = 7.2% return. But if early repayments exceed 20, returns are back to zero.

  4. #2584
    Guru
    Join Date
    Aug 2012
    Posts
    4,663

    Default

    Quote Originally Posted by leesal View Post
    Agreed



    Hazard Curve


    The forecast default rates are shown as a consistent annual rate over the term of the loan. In reality, defaults are more likely to follow what is known in statistical terminology as a time-varying hazard rate. This chart shows the profile of the hazard curve of the personal loan portfolio to date and shows that almost 60% of the defaults that have occurred may been within the first 15 months of the loan.
    Plus defaults are not tax deductible for investors not in business, so the effective tax rate on the blended return (including charge offs) of high risk notes could be well over 33%.

  5. #2585
    yeah, nah
    Join Date
    Mar 2017
    Posts
    491

    Default

    Quote Originally Posted by leesal View Post
    If Harmoney's marketing is to be believed, 50% of the defaults occur within 12 months. At grade F5 that gives- 24%. So in the simplistic model 24 gone and 76 paying 30% interest 23%. ~ negative (1%) return.

    Year 2 resets itself- 25% more of total defaults - 12 gone, and remaining 64 paying 30% (19.2) = 7.2% return. But if early repayments exceed 20, returns are back to zero.
    I think the numbers you are using are very much in favour of showing a larger loss than they should:

    Based on 9.49% pa default rate over 5 years the total default rate is actually 32.89% not 47.5% as you suggested above - the 9.49% needs to be applied to what remains, not the original amount each year.

    Taking your suggested 50% at 12 months, that's 16.45%, not 24%, the end result will be somewhat different to what you have shown?

    Year two, if you take another 50% of what's left as you have, it would not 'reset' itself, it can only be of what is left of expected defaults, which at 50% would be 8.23%, further reducing as it goes... If you focus on the front end losses you miss the fact that you make more gains towards the end of the loans (if they don't get paid out, but if they do that is a gain).

  6. #2586
    Member
    Join Date
    Mar 2017
    Posts
    44

    Default

    Is it just me, or has loan volume seemed to have increased since the new scorecard was implemented?

  7. #2587
    Member
    Join Date
    Aug 2017
    Posts
    212

    Default

    Quote Originally Posted by myles View Post
    I
    Based on 9.49% pa default rate over 5 years the total default rate is actually 32.89% not 47.5% as you suggested above - the 9.49% needs to be applied to what remains, not the original amount each year.

    Taking your suggested 50% at 12 months, that's 16.45%, not 24%, the end result will be somewhat different to what you have shown?
    Indeed it would

    Although I get a different result, 39.26%. Are you feeding in early repayment

    =9.49+(1-(def+er))*9.49+(1-(def+er)^2)*9.49+(1-(def+er)^3)*9.49+(1-(def+er)^4)*9.49
    where er = 9 gives 32.89%
    Last edited by leesal; 23-08-2017 at 12:07 PM.

  8. #2588
    Member
    Join Date
    May 2016
    Posts
    236

    Default

    Quote Originally Posted by leesal View Post
    Indeed it would

    Although I get a different result, 39.26%. Are you feeding in early repayment

    =9.49+(1-(def+er))*9.49+(1-(def+er)^2)*9.49+(1-(def+er)^3)*9.49+(1-(def+er)^4)*9.49
    where er = 9 gives 32.89%
    I don't think it is as straight forward as that. I would think the default rates would be assessed considering the typical expected life of loans rather than just straight lined average over total life of full term loan. In other words the Hazzard curve is partly allowed for in the annual default rates (ie they are increased). This is borne out by the fact actual defaults are pretty close to expected.

  9. #2589
    Member
    Join Date
    May 2014
    Posts
    204

    Default

    Quote Originally Posted by leesal View Post
    How can F5 possibly be profitable?

    According to the forecasts, annual default rate of 9.49 annually, applied over 5 years gives a loan default rate of 47.5%. The hazard curve then suggests So 24% default against 30% interest in the 1st year.

    2nd year and onwards, a severely eroded premium base (even without early repays) , so fall short of recouping default losses through remaining cashflows.

    Assuming it takes 6 months for the default to register, ie it goes 6 months without it paying you anything, then its defaulted.
    Then for year 1, on average...
    90.5% of the time you collect a full years interest. 9.5% of the time, you collect sum or no interest (average would be 3 months interest i think).

    Year 2, you have got a little bit of capital back, maybe the loans 90% of its original value. so again, 90.5% of the time, you get a full years interest (obv less than year 1), 9.5% of the time it defaults, and you get not as much interest back

    Extrapolate this out over a sample of F's, and its profitable.

    Model a 30% interest rate on a $100 loan, you only need to last 31 months (out of 60) to get all your $100 back (in nominal terms).

  10. #2590
    Member
    Join Date
    Jul 2017
    Posts
    128

    Default

    Quote Originally Posted by CageyB View Post
    Is it just me, or has loan volume seemed to have increased since the new scorecard was implemented?
    Yes definitely more loans available - the lower interest rates are very attractive. 6.99% over 5 years for an unsecured loan is exceptional - even better than Heartland Bank's secured mortgage rate! I think that we the lenders will earn a lower RAR over time but Harmoney will be winners with many more $500 lending fees earned funded up front immediately by us?
    Last edited by joker; 23-08-2017 at 04:44 PM.

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •