Endowment Compensation - Which interest rates to use in compensation calc?

I have had a complaint upheld and the lender is going to calculate the compensation. To do this they have said that they will either use their standard rates over the period or our actual rates (we changed lenders a couple of times always using the endowment as a repayment vehicle). The thing I can not work out is that if the interest rates from our actual lenders were say lower or higher than the lenders standard rates would it be better to go with their rates or the actual rates.

Any help from anybody who understands this would be most appreciated.

Reply to
tom
Loading thread data ...

In message , snipped-for-privacy@tungsten.demon.co.uk writes

The lower the interest rate the quicker the debt is reduced in the early years, in fact a graph showing the reducing balance becomes straighter the lower the rate. The higher the interest rate then the slower the debt gets repaid in the early years and the graph becomes more skewed towards the top right hand corner, i.e. more curved.

Reply to
john boyle

The lender? Surely you mean the endowment company, who may also be a lender when wearing a different hat, which is presumably where their "standard" rates come from, unless they just use a "typical" market cross-section.

The endowment provider will presumably beef up the size of your endowment surrender value to equal the amount you would have paid off your mortgage loan, had you opted for a repayment mortgage from the outset.

If there are no complications which changed the value of your debt, such as you making lump sum overpayments, taking further advances, or changing houses, and also assuming (as no doubt they will) that when changing lenders you would have (had you had a repayment deal) kept the projected end date constant, namely the same as that of your endowment maturity date, then it is the case that the lower the interest rate, the faster you'd be paying off the capital.

So if your actual lenders' rates were always lower than the standard, you would have paid the debt off quicker, and so you would be better off choosing the actual rates, and conversely if your actual lenders' rates were always higher than the standard, you would have paid it off more slowly, so you'd be better off choosing the standard rates.

If, however, your rates were sometimes lower and sometimes higher than the standard, it may balance out.

For example:

First, remember that all lenders' rates, including the "standard" rate, will have fluctuated in line with BoE base rate, except where fixed rates apply. But this illustration, for clarity, assumes no fluctuation.

Suppose the "standard" rate is 5%, and consider higher and lower rates of 6% and 4%. After 5 years of a 25 year term, you would have paid off

13.01%, 11.58%, 10.27% of the original loan (at rates of 4%, 5%, and 6%, respectively).

But if you were on 4% for the first 5 years and then on 6% for the next

5 years, it pretty well cancels out. At 5% for 10 years you'd have paid off 26.35%, but at 4% for the first 5 and 6% for the next 5, you get to 26.34%. The other way round, 6% for the first 5, then 4% for the next 5, gets you to 26.59%.

But this is only half the story. They will presumably also make an adjustment for the cost of your payments. Whatever rate they use, you will have paid off more with the repayment plan than with the endowment and the interest-only loan. But you would have spent more doing so, assuming that the extra cost of a repayment payment over the interest-only payment is less than the excess of the endowment payments over the cost of an equivalent term life policy.

What makes things tricky is that the excess of a repayment payment over an interest-only payment is smaller when the interest rate is higher, and vice versa. So although at 5%, after 5 years, you'd have paid £11,580 off a £100k loan, it would have cost £35,476 in payments, £10,476 more than interest-only. At 6% after 5 years you would have paid off £10,270 (£1,310 less than for 5%), and it would have cost you £39,113, or £9,113 more than IO (£1,363 less than for 5%, which more than cancels out the disadvantage of having paid off less). At 4%, you'd have paid off £13,010 (£1,430 more) at a cost of £32,006, or £12,006 more than IO (£1,530 more than for 5%, which more than cancels out the advantage of having paid off more.).

So, to conclude, go for whichever rates are generally higher.

Reply to
Ronald Raygun

Entirely true, but that costs comparison leaves out some of the payments that are taken into account.

The costs comparison includes: Endowment mortgage: interest payments to lender *and* endowment policy premiums Repayment mortgage: payments to lender *and* the cost of a decreasing term assurance policy (DTA) to protect the mortgage (where applicable, and it generally is).

In general this means that the endowment and repayment method cost about the same. If the cost of a DTA is not being taken into account, then I would say you want to go for rates that are generally lower. If it is, then I think its very close - I've seen plenty of calculations that gave higher compensation using generic rates than the customers' actual rates, and vice versa.

In essence, if you ask the firm to use the rates you have actually paid, then you will get exactly the compensation you are due. If they use generic rates, then you may get back more or less than the amount that should be due.

Laura.

Reply to
Laura

Yes, that's true enough, but misses the point at issue, I think.

The size of the endowment premiums and of DTA premiums are fixed at the outset, and are independent of the loan interest rates. Therefore they can be disregarded when comparing the effect of using one set of loan interest rates against another, since they will affect both scenarios equally.

Agreed, and furthermore, we're not talking about the difference between one and the other, but the difference between (the difference between one and the other when using one set of interest rates) and (the difference between one and the other when using another set of rates). All in all, the bottom line difference is likely to be almost too small to bother about, and the hassle of digging out a complete history of actual rates is outweighed by the convenience of letting them use their generic rates which are already available to them. Only if those rates are somehow rigged in their favour would it be unwise to trust them, but you're trusting them with the actual calculation anyway, aren't you? How much detail do they tend to provide to allow you to check their workings, to let you be convinced they're not diddling you, bearing in mind that the whole compensation situation is the result of them, in a way, having in fact diddled you?

Perhaps they should be made to let you provide your actual rate history, and to do the calculation twice, once with their rates and once with yours, and pick whichever gives you more.

I don't understand why you say that. It should affect the difference, but not the difference difference.

Exactly, but what the OP wanted to know is which of more or less it is, based on whether the actual or generic rates are generally higher or lower.

Reply to
Ronald Raygun

Fair enough, but I understood you to be saying that, with a lower interest rate, the increased reduction in the mortgage balance of a repayment would be cancelled out by the excess of a repayment payment over an interest-only payment. I'm not convinced that the excess in the payments to the lender can be compared with the reduction in the mortgage balance unless the total cost of each mortgage option has also been taken into account.

Also, the endowment premiums aren't necessarily fixed at outset - a number of unit linked policies include reviews that may increase the premiums at certain policy anniversaries. This clearly doesn't affect a comparison of an interest only mortgage with a repayment, but it can affect an RU89 calc.

They tend to give you (if you ask) a printout which shows the endowment premiums paid, DTA premiums assumed, lender, term, initial mortgage amount etc. Most firms use a software package called Mortgage Fundamentals, and this will also give you a printout of the interest rate used for each month along with each month's mortgage payment. It's a bit of a pain to trawl through, but it can certainly be done.

Well yes, but that would result in some people being overcompensated - I can't see the FSA being particularly keen on that option, as firms would end up having to compensate people for losses that they haven't acutally suffered. The option of generic rates is only there for administrative convenience for everybody, but some people will be undercompensated if they use this method.

The rates used can make a fairly significant difference, especially on a large mortgage where the interest rate has been fixed for a long period (and can therefore be some way off SVR).

I said that because I'm being dozy - sorry! I'm not quite sure what you mean by the 'difference difference' though?

I still think that the answer to this depends on the actual cost of the endowment premiums and DTA premiums. However, if the rates the borrower has paid have been pretty close to SVR, sometimes higher and sometimes lower, as you say it won't make much difference anyway, unless he has had a long term fixed rate or has done something non standard with his mortgage. Deferred interest mortgages and such like are another kettle of fish, and if the OP's mortgage has been arranged in this way a calculation with generic rates is highly unlikely to be appropriate.

Laura.

Reply to
Laura

Yes, that's roughly what I meant.

Then I must try to convince you.

I've never heard of such, but I'll take your word for it that they exist. My point is, though, that the difference difference calculation makes any premium variations drop out, since they are the same in both scenarios being compared. They don't vary with choice of interest rates.

I don't see anything more wrong with that (the providers are, after all, in a sense being punished for mis-selling and it seems not inappropriate that they should over-compensate) than with them giving the customer the option to gamble between being over- and under-compensated should they elect to choose generic as opposed to actual rates. After all, most customers don't have the wherewithal to work out whether going generic will gain them something or lose them something, so it is in most cases a genuine gamble, unless they pay someone knowledgeable to run through both calculations for them, in which case the cost of this advice may outweight the benefit of know which is the better choice.

OK. By plain one-word difference I mean the difference in the client's financial position at a time N years into the term, between on the one hand having opted for the endowment method, and on the other for the repayment method. Those two financial positions being measured as, in one case, the value of the endowment fund, minus the total paid (being

12xN endowment premiums plus 12xN interest-only payments), and in the other case, the value paid off the initial debt, minus the total paid (being 12xN DTA premiums plus 12xN "normal" loan repayments).

But we really have four scenarios we need to compare, namely: (1) endowment with interest-only loan at actual interest rate, (2) DTA with repayment loan at actual interest rate, (3) endowment with interest-only loan at generic interest rate, (4) DTA with repayment loan at generic interest rate.

So one plain one-word difference would be (value minus total cost of scenario 1) minus (value minus total cost of scenario 2). A second one-word difference would be that involving scenarios

3 and 4 instead of 1 and 2.

By two-word difference difference I basically mean the difference between these two differences, i.e. between the scenario-1-2-difference and the scenario-3-4-difference.

The fact is that scenario 1 is the only one which actually happened, the other three scenarios are fictitious, the "correct" compensation, we agree, is based on comparing scenarios 1 and 2.

But if they let the client choose generic interest rates instead, then we should not be comparing scenarios 1 and 4, but 3 and 4, since it is only fair that if they pretend that the fictitious repayment loan involved generic rather than actual rates, then they should likewise pretend that the generic rates applied to the interest-only loan as well. Perhaps they don't in fact work this way, but I've been assuming they do, because it just doesn't make sense any other way.

So, the difference to the total compensation payment, due to the choice between actual and generic rates, should be equal to my "difference difference", because if you choose actual rates, your compensation should equal (Value2-Cost2)-(Value1-Cost1).

[Oh heck, this is already a difference difference, perhaps I should have said "difference difference difference"]

And if you choose generic rates, the compensation should equal (Value4-Cost4)-(Value3-Cost3).

What the OP was interested in is which of the above is bigger, i.e. we're interested in the sign and magnitude of the expression

((v4-c4)-(v3-c3)) - ((v2-c2)-(v1-c1))

Now, we know that v3 and v1 are the same, they are the value of the endowment fund, which is independent of the interest rate chosen. Since it appears on both sides of the main minus sign, it can be removed from both sides without changing the result, and we get

(v4-c4+c3) - (v2-c2+c1)

We also know that c1 and c3 are the sum of the costs of the endowment premiums "e" and of the cost of servicing the interest-only loans:

c1 = e+ioa (ioa stands for interest-only, actual) c3 = e+iog (iog stands for io, generic)

and that c2 and c4 are the sum of the costs of the DTA premiums "d" and of servicing the repayment loans.

c2 = d+ra c4 = d+rg

What does that do to our expression? It becomes

(v4-d-rg+e+iog) - (v2-d-ra+e+ioa)

and as you can see both sides contain the same -d+e so they cancel out. All we're left with is

(v4-rg+iog) - (v2-ra+ioa)

which I'd like to simplify to (v4-xg) - (v2-xa) where xg (or xa) is the excess of the cost of the repayment-loan payments to date at the generic (or actual) rate over the cost of the corresponding interest- only payments.

QED. Convinced yet?

Reply to
Ronald Raygun

"Ronald Raygun" wrote

Surely the option for using their standard rates is there because it may be difficult to obtain, or calculate, using the actual rates. The standard rates therefore are simply an option to either save time/expense in calculation, or because the actual rates are just plain not available.

Your suggestion to do *both* calculations assumes that the actual rates

*are* available, and that it *is* possible to produce the calculation using them. In that case, there should be no need (and indeed no option) to use standard rates - everyone should have their reviews performed with the "correct" *actual* rates.

Bearing in mind that "actual rate" data had been obtained, and a calculation had been performed using these - what, do you think, would be the purpose of then *also* using standard rates for a similar calculation??

Reply to
Tim

Thanks Ronald and Laura for some fascinating info. Way above my head but still it highlights how complicated it all is (and how easy it would be for a lender diddle someone - allegedly of course).

The actual scenario I find myself in is that the lender (yes it is the lender - I thought this was strange as they are not the endowment provider however they have done all the communications) has offered to use their rates to calculate the period when the mortgage was with another lender or the actual rates of the other lender. I have detailed their rates as follows;

1st July 98 - 8.95% 1st Nov 98 - 8.70% 1st Dec 98 - 8.20% 1st Jan 99 - 7.70% 1st Feb 99 - 7.45% 1st Mar 99 - 6.95% 1st May 99 - 6.99% 1st Oct 99 - 6.99% 1st Dec 99 - 7.24% 1st Feb 00 - 7.49% 1st Mar 00 - 7.74% 1st Mar 01 - 7.50% 1st May 01 - 7.25% 1st Jun 01 - 7.00% 1st Sep 01 - 6.75% 1st Oct 01 - 6.50% 1st Nov 01 - 6.25% 1st Dec 01 - 5.75% 1st Mar 03 - 5.65% 1st Aug 03 - 5.50% 11th Nov 03 to date - 5.25%

Over this period my actual rate was fixed at 5.9% from 1st May 98 to May 02. It then went on to the standard variable rate for the lender for the period May 02 to May 03. I have lost my records of this rate but I seem to remember the payment went down by about £150 from when the rate was fixed so I think it was around the 4.5 - 5.0% range. In

97 I did increase the loan, the increase being on a repayment basis, and extended the period by about 5 years however a portion of the loan (£60K - this was the original mortgage loan amount) was still secured with the endowment.

The original mortgage was taken out with the lender in Mar 89 and was interest only until I remortgaged in June 97. At this time I switched the whole loan on to a repayment basis but kept the endowment going. At a later date on the suggestion of a financial adviser from another BS I remortgaged again, secured part of loan with the endowment (£60K) and had the rest as a repayment, all fixed at 5.9% for 4 years.

In Dec 03 I remortgaged again when I found that I had a £35k shortfall, coverting the shortfall to a repayment mortgage and fixing the whole loan at 4.25% for 5 years. This had some arrangement fees attached to it which I think if I understand the FSA guidance notes correctly I can claim as well.

All this time the endowment premiums have been constant at £78/month.

One other thing that I did was pay a lump sum of £11K in to the original mortgage account to reduce the loan capital. This was done to get out of a bad negative equity trap I was in (that's whole other story I like to keep locked up at the back of my mind labelled 'nasty experiences for the naive first time buyer').

With all this complication I get the feeling that I haven't got a chance in hell of making sure the compensation I get is right. Any thoughts you have would be most welcome though.

Many thanks,

Tom

Reply to
tom

That's one reason, but the cynic in me won't believe it's the only one.

Indeed it does.

The way the story was presented was that the customer was given the choice before it had even been established whether the actual data were available. Obviously if they were not, the choice would be removed and the standard rates would have to be used.

To determine which set is more beneficial to he customer, of course. Silly question.

I mean, if asked "Do you want us to use your data or our data?", what should one answer? "Er, er, I don't know, you choose." or "Sorry, I have no data, so I guess we'll use yours" or "Hang on a sec while I toss this coin" or "Well, what *are* your data? Let me see. Hmm, yes, OK, let me get back to you in a couple of days. Hello? It's me. You know, the rates data? Yes, I've decided. Let's use mine!" but the most likely first reaction would surely be "Er, why do you ask? What difference does it make?", to which the company can hardly answer "Well, you may get more money with one than the other, but we won't help you guess which is which.", can it?

Reply to
Ronald Raygun

Ronald Raygun said: "What makes things tricky is that the excess of a repayment payment over an interest-only payment is smaller when the interest rate is higher, and vice versa. So although at 5%, after 5 years, you'd have paid 11,580 off a 100k loan, it would have cost 35,476 in payments,

10,476 more than interest-only. At 6% after 5 years you would have paid off 10,270 (1,310 less than for 5%), and it would have cost you 39,113, or 9,113 more than IO (1,363 less than for 5%, which more than cancels out the disadvantage of having paid off less). At 4%, you'd have paid off 13,010 (1,430 more) at a cost of 32,006, or 12,006 more than IO (1,530 more than for 5%, which more than cancels out the advantage of having paid off more.)."

I'm afraid I got lost in all the calculations, so I'd like to use your actual figures to explain where I think my confusion is! I think it is related to the fact that notional savings are not generally taken into account, but I'm not sure.

Say we do have a 100k loan, and we are doing the calculation after 5 years.

If interest rates were 4%, the balance of the loan would have reduced by

13,010, but repayment cost 12,006 more than interest only. If we say that the endowment premiums totalled 12,006 (ignoring DTA) and that the sv of the policy is 10,270, then our complainant has made a loss of 2,740. (That is, his loss is the difference between the balance of the loan and the sv of the policy).

If interest rates were 6%, the balance of the loan would have reduced by

10,270, but repayment cost 9,113 more than IO. If we say that the total endowment premiums were 12,006 (ignoring DTA), and the sv was 10,270, then our complaint is 2,893 worse off. (That is, his loss is the difference between the cost of the repayment mortgage and the cost of an endowment mortgage).

So far, he would be better off to choose higher rates.

I don't contend that the *absolute* value of the endowment premium makes a difference (although I appreciate I have earlier said just that), but I still think that the difference between the endowment premium and the DTA premium makes a difference.

To take an extreme example, say the premium for the endowment equals the premium for the DTA. (With very small policies where both premiums would be the minimum available, this is possible). I'm also going to assume that the surrender value of the policy is 10,270.

If interest rates were 4% per annum, our complainant gets redress of 2740. (That is, the difference between the sv of his policy and the amount by which a repayment mortgage would have reduced). He is not due any redress with respect to his outgoings, because he has saved 12,006. These savings will generally not be taken into account.

If interest rates were 6% per annum, our complaint would get redress of 0. The sv of his policy is the same as the amount by which the balance of a repayment mortgage would have reduced.

Here our complainant is better off with the lower rate of interest.

I hope I've made myself clear! Are we just disagreeing about whether notional savings are taken into account, or have I got the wrong end of the stick entirely? Or have I just run away with somebody else's stick...

Laura.

Reply to
Laura

Thanks for taking the trouble to clear this up.

OK.

Agreed.

Well, yes, if by "balance of loan" you mean "amount paid off loan". More generally, the complainant's loss is:

(amount paid off - extra cost of rep) - (SV - premiums)

which itself is a simplification of the even more general

(amount paid off - cost of rep) - (SV - cost of IO - premiums)

But in your example the "extra cost of rep" happens to equal "premiums", so here loss happens to equal (amount paid off) - (premiums).

Agreed.

Indeed, but as above, this is only so by coincidence (or rather due to your cunning choice of SV and premium cost). More generally, as above, his loss is the difference between the profit he would have had with repayment (£1157) and that he in fact had with the endowment (-£1736). That difference is £2893.

Yes, at 6% he is £2893 worse off, while at 4% he is only £2740 worse off. Since his compensation will equal the amount by which he is worse off (or is deemed to be worse off), he should choose the rate at which he is worst off. He gets £153 more compensation if he chooses 6% than if he chooses 5%.

We're getting closer to pinpointing where we are failing to connect (that's a diplomatic way of saying "where your misunderstanding lies"). I don't understand why you think so. I say this difference is indeed relevant in determining the absolute size of his redress, but as it does not vary with the assumed interest rate, it will not affect the

*relative* size of his redress when he chooses which interest rate to use in the redress calculation.

I don't think that's a helpful assumption, but I'll humour you.

No.

Wait! There is nothing special about 4%, except that with your figures in your ignoring-the-DTA example, the cost of the endowment premiums happened to match the extra cost of REP vs IO loan payments. You cannot generalise from an observation in your earlier example, which was coincidental, to what happens if you change a paramater.

In this case, in fact, the complainant gets no redress, because he has not made a loss, he is in fact £9,266 *better* off with the endowment. This is because in your earlier example, the 4% repayment made a profit of £1004 while the endowment made a £1736 loss, so the endowment made him £2740 worse off than the repayment. In this later example, though, using the same numbers but adding the cost of DTA, the repayment method now has the added £12006 cost and as a result the £1004 profit becomes a £11002 loss, while the endowment loss is unchanged at £1736. So the endowment makes him £9266 better off.

The "redress" would be negative, and so there is none.

He has actually saved £9266, but yes, because he has saved instead of lost, he gets nothing.

Not quite. With the repayment, he would have made a loss of £10849, that is, the £10270 paid off, minus the £9113 extra cost of the loan, minus the £12006 extra cost of DTA. Again, the repayment loss would exceed the endowment loss (this time by £9113), so again he would get no redress (as you say, but for a different reason). He would get no redress because he would have been worse off with repayment.

I'm afraid not. In both cases he'd be equally well off and get no redress either way, because in both cases he will have been better off with the endowment (since they actually produced a positive SV) than with DTA costing as much as the endowment.

I'm not quite sure what you mean by "notional" savings. In my view the DTA cost is as important to take into account vis-a-vis the endowment premium cost as is the IO loan servicing cost against the repayment loan cost. It seems sensible (as we have done) to ignore the common part of the IO cost and only list the excess of repayment over IO against the repayment method figures. It should therefore be equally sensible to list only the excess of the endowment premiums over DTA premiums against the endowment method figures, because despite your extreme example, generally EP cost will exceed DTA cost. The important thing is that neither the endowment cost nor the DTA cost, nor therefore the difference between the two, changes with loan interest rates, and so will always be the same.

Reply to
Ronald Raygun

BeanSmart website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.