Central Bank letter on Structured Products simply ignored

D

Duke of Marmalade

Registered User
Messages
4,686
In April, the Central Bank issued a Dear CEO Letter to firms producing Retail Structured Products (RSPs).
This is what they said on "back-tests" which are ubiquitous in the brochures of these complex products.
You might have though that this would see an end to this obscene misrepresentation. Well I cite 4 examples below of RSPs issued recently involving 3 "master brokers".
Product A
The Key Information Document (KID) tells us that in 10% of forward looking simulations there would be losses of at least 42% of initial capital. The KID information is not presented in the marketing brochure. Instead we are told that 3,815 daily overlapping back-tests produced zero instances of a capital loss.
Product B
KID tells us 10% of forward simulations would lose at least 24% of initial capital. Brochure tells us that 3,319 back-tests produced zero instances of a capital loss.
Product C
KID tells us 10% of forward simulations would lose at least 98% of initial capital. Yes that's right 98% loss in 10% of situations! Brochure has 1,304 back-tests producing zero instances of loss.
Product D
KID tells us 10% of forward simulations would lose 10% (this is not a kick-out bond, and 10% is the maximum loss). The brochure shows that 4,518 back-tests had zero instances of loss.

What happened to the Dear CEO Letters sent to the CEO's of these firms? Did they just make paper darts of them?
 
Last edited:
I have relied on backtesting historically when speaking with brokers or advisors but the above suggests it's not be relied on. I didn't understand why tbh.

a quick google found this article from 2013 & I think the below extract explains the issue.
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://newfrontieradvisors.com/media/1450/2-7-13-be-cautious.pdf

"Moreover, a major drawback to backtesting is the fact that past time periods can be arbitrarily selected and data easily manipulated to suit a desired outcome. For instance, a proposed strategy could be adjusted to perform well in a particular time period, say 2008 to 2009, or conversely, a specific time period could be adjusted to show success for the strategy. The manipulated strategy can then be refined repeatedly to what the strategists call “a perfect fit” to a historical period. However, what has actually been created is a customized strategy that performed well for a particular investment period, but it in no way ensures similar performance for the future. In other words, refining a backtest until a desired result is achieved can be extremely dangerous or risky, and at the very least, is a dishonest practice. "
 
@Horatio indeed, but in this current situation there is no problem picking an investment period. Any product tested over the last dozen years or so of bull market would succeed. For a 10 year product there has really been little more than one observation of how it would have done in the past. But one back-test would hardly set the brochure alight. So they manage somehow to have, say, 3,000 back-tests performed in the recent past. Whilst 1 observation wouldn't carry much weight 3,000 observations of essentially the same investment backdrop sounds oh so convincing. How do they do it? They simulate a product being launched every day. Of course it is not practical to do that. That is ok, simulations do not have to be practical. But why stop at daily overlaps? The data is there to perform minute by minute simulations. That would give 3 million successful back-tests. They face the snake-oil poser of where to draw the line. A snake oil salesman will try and persuade you that his product will make your hair grow back but he will baulk at trying to convince you that it will make you the heavyweight champion of the world. Daily overlaps are at the "hairgro" frontier of the back-test snake oil.
 
Last edited:
Here's another quote from the Dear CEO Letter.
A fixed decrement index? Well ostensibly the decrement is a proxy for dividends. The underlying index is what is called Total Return, that is it includes dividends. So the fixed decrement is used to knock out the dividends. Oh we are told that the fixed decrement will likely exceed dividends. But that is not the main trap for trusting advisors and their clients alike. The key word is "fixed". So for example we might have the underlying index starting at 100 with a fixed decrement of 5. Looks a bit like 5%. But if the index falls to say 50 it is now 10%. True that if the index rises to 200 the effect is 2.5%. So a sort of balancing symmetry. Except the products that use these fixed decrements chop off the upside but leave full exposure to the downside. Thus the punter never will see the benefit of the 2.5% charge, it "kicks out" way before then, but the 10% charge can be very real as it tips her over the precipice where she suffers the full downside.
A broker friend of mine emailed me when the Dear CEO Letter was published to proclaim "thank goodness that's the end of fixed decrements". How innocent he was. Product A above is based on a fixed decrement index. At time of printing the brochure the underlying index was 82.2 so the going rate of the fixed decrement of 5 was 6.1%. Product B is also a fixed decrement product with the current value of the index 80.745 and a corresponding current decrement % of 6.2%.
 
Last edited:
It is heartening to see that Financial Advisors contributing to this thread are as disgusted with the CB impotence as I am. I am also receiving similar support in private emails from Financial Advisors (who are of course like me members of the aristocracy).
It would be great if Financial Advisors who feel as strong on this point as I do dropped a short email to their Central Bank contact referring to this thread and simply asking "can this be true?".
 
Initially, I thought it was overkill by a provider to show zero failures in over 2,000 "back-tests". Why not show (say) 2 failures when the true forward-looking risk of failure is one in six? Then I realised that even admitting explicitly to a 1 in 1,000 risk of failure could destroy sales prospects. Potential buyers would be frightened off if they realised there was even a 1 in 1,000 chance of losing at least half their capital, possibly much more, given that they signed up in the belief that they would get a bit more interest than from the bank.
The CBI doesn't prevent providers from producing brochures that show zero failures in back-tests, and which give no indication of prospective risks, so we must presume that they see all this as complying with MiFID regulations, which oblige investment firms to ensure that ALL (not "some") marketing communications are fair, clear and not misleading. They didn't reply when I asked them a straight question to that effect.
As the Duke suggests, others should ask the same straight question of the CBI: do they think that brochures which show zero failures in back-tests, and no indication that the forward-looking risk of losing more than half one's investment is around 1 in 6, are complying with the MiFID requirement that ALL information in sales literature is "fair, clear and not misleading"? Maybe even a few politicians could be primed to ask the same question. Would the CBI ignore them as well?
 
My Central Bank contact?? We don't get to talk to the Central Bank! Everything is through their incredibly difficult to use online system.
 
You must log in or register to reply here.
Menu
Log in
Register