FSA chutzpah on structured products
Though promoted by regulatory developments which are likely to be mainly of interest to our peers in private wealth management and financial planning, our explanation in this post of how we think suitability of structured products for individual clients can be assessed is relevant to both professionals and serious private investors who have used structured products, as alternatives to either deposits or unhedged risky asset exposures. We welcome observations if readers disagree with our analysis.
It can be used by financial journalists as background briefing on structured product issues as well as specific commentary on the FSA’s review of weaknesses in advice processes.
The specific regulatory issues we highlight in this post tie in with a No Monkey Business theme (originally aired in my book) that UK financial services regulation has consistently failed the public by focusing on processes at the expense of products, a message we hope both peers and the media will support. It is advisers with an unblemished record who end up paying every time the product manufacturing industry screws up and I am sure we are not alone in running out of patience with our regulator on this one.
Another case of missing product regulation.
Two recent publications by the regulator suggest it has never understood the consumer protection issues around structured products which have limited or complete capital protection and use options to provide the return potential.
Last week the FSA published a report of its investigations of weaknesses in advice processes when firms promoted structured products with capital guarantees. This investigation follows the collapse of Lehman, Keydata and Arc.
Reviewing about one third of the firms that were heavy promoters of structured products to private clients, the FSA found the customary failings in record keeping that crop up in any review when product distribution goes wrong. But we are mainly interested in those it singled out that are specific to these products: faults in suitability assessments and risk disclosures, notably widespread failure to explain the counterparty risk associated with the capital guarantee. There is not much sympathy for the fact that complex and innovative products are bound severely to test advice and distribution processes and systems.
In a different context the FSA addressed structured products in June, with the publication (CP09/18) of its final consultation paper on the Retail Distribution Review. In its guidance to its proposed draft rule COB6.2.A10 G for testing whether an adviser’s status in a ‘relevant market’ is ‘restricted’ or ‘independent’, the FSA said this (2.8):
‘One of the challenges for independent advisers will be to ensure they have sufficient knowledge of all types of products which could give a suitable outcome for their clients. The rules do not mean that we expect to see all advisers recommending products such as structured investment products, for example, as a matter of course. But we would expect that if a structured product would best meet the client’s needs and risk profile, then an independent adviser should have sufficient knowledge of these products to be able to recognise this and make a recommendation to buy this product.’
Along with many other wealth managers and independent financial advisors, we were appalled that the FSA seemed to be arguing that structured products were to be considered prima facie as a suitable solution in a retail market where most (if not all) alternatives have no principal risk (or risk is practically eliminated by diversification) and have no optionality or asymmetric payoffs.
Principal risk and optionality are the two key attributes that raise consumer protection issues for product structures that are positioned in the market as low-risk. As my book pointed out, we have a long history in the UK of product failures where the products were designed to improve returns relative to simple deposits while disguising the risks. If it was true in 2002 that consumer protection regulation should have heeded the Gower Commission’s original recommendation to focus on products not sales processes, it is surely now self-evident that we would have avoided many if not all of the problems caused by product innovation by clever investment bankers. It is of course patently true of banking markets, where the FSA failed to anticipate problems caused by the structured loan products.
The Treasury is also to blame as it has used National Savings & Investment’s Guaranteed Equity Bonds as a source of cheap finance without regard to the public’s poor grasp of option bets.
Counterparty risk There is no excuse for a distributor failing to understand or point out the risk of a counterparty failing, whether it is as simple a product as an annuity or as complex as capital-guaranteed or capital-at-risk structures.
That said, I am sure the risk of a principal failing does not always get attention drawn to it when an annuity is sold, just because it is several decades at least since an insurance company failed to meet its liabilities to annuitants and centuries since insurance companies took over writing annuities from governments after the latter started defaulting on their obligations. Advice about principal risk also looms as a major problem for the FSA in respect of final salary schemes, when pension benefits exceed the protected level of just £30,000 pa. There is an overwhelming case in finance theory for reducing this uncompensated idiosyncratic exposure but also a very strong regulatory bias against advising transfers.
The nature of capital guarantees is particularly opaque in structured products. The products were created and distributed by issuers who are often household names but many buyers are (or were) probably not aware that the guarantee is not usually provided by the issuer but by a completely separate commercial bank or investment bank, often less well known than the issuer and in many cases located abroad. I believe in some cases the identity of that counterparty was not even disclosed in the product literature. It needed no special insight by the FSA to see the case for explicit and prominent exposure warnings being prescribed in the product literature , so as not to be dependent on the adviser’s sales process.
Principal risk is managed by exposure limits or diversification. But (unlike equity and bond markets) it is not practicable in product markets (or contractual savings that are not completely covered by compensation schemes) to diversify all the specific principal risk. It is also unreasonable to expect consumers to avoid all principal risk. It is necessarily therefore a matter of judgement what exposure levels should be retained.
If this is a risk properly managed by exposure limits, what is to stop almost anyone from complaining about the sale of a structured product (or perhaps several based on different underlying risk assets, with different issuers) which exposed the customer to ‘excessive’ credit risk on the same counterparty?
Unless there is a systemic review process like pension mis selling, all such unresolved complaints have to be handled individually by the Financial Ombudsman Service. Even when systemic patterns emerge, as has been the case (for instance) with the widespread sale by banks of non-deposit based investment products to their own risk-averse depositors, the FOS cannot avoid its obligation to consider every case on its own merits.
What, then, should be the FOS’s basis for assessing appropriate levels of undiversified credit risk exposures to financial institutions and how will it ensure it commands the respect of both the industry and the advisers? If it looks at each individual’s past experience of holding uninsured credit exposures, which is a reasonable basis, the chances are (like my annuity example) it will find no precedents to use because so few packaged products introduce this risk.
The FSA review hardly refers to the attributes of option-based structures that pose particular problems for assessing suitability of advice. Either they are being disingenuous or they do not understand the issues. The comments in the RDR consultation paper quoted above suggest the latter.
One of the advantages of product regulation, whether as a licensing process before innovations can be marketed or as thematic reviews when new products start to be widely marketed, is that it requires structured products to be looked at from the manufacturing point of view, breaking the structure down into its active ingredients. Many of the structural features will turn out to be common to a broad range of structures and, if there are marketing issues posed by them, they will be common to all.
The typical five-year ‘capital guaranteed’ equity index structure, for instance, can be emulated by, or actually consists of, a five year discounted bond plus a five year call option on an index which is the underlying assets on which the option is taken out. It is only because the bond is discounted (as in £80 invested pays back £100 at maturity) that there is cash available from the total investment to pay for the option and the structure costs. The payoffs of the structure can be replicated with swaps or put options but the easiest way to understand the structure is a discounted deposit combined with a call option.
It should be clear that the if the investor would not, outside a product structure, choose to apply the money represented by the deposit interest to the purchase of an illiquid call option, then clearly it is difficult to justify the product sale. Unfortunately, that is not how the choice was presented.
If the full amount of the invested capital is guaranteed by a counterparty, then only in the narrow sense of staking income rather than capital can the payoff impose an investment loss, as opposed to loss caused by the failure of the counterparty to honour its obligations.
If the capital is underwritten at less that 100%, it is termed a SCARP, or structured capital at risk product. Nowadays, capital at risk is typically limited to about 10%. The purpose of risking a bit more capital is to provide more optional upside participation. There is nothing fundamentally different between a capital protected and capital at risk product except in terms of this narrow distinction between capital and income.
(This is different, note, from the SCARPs that blew up spectacularly in the 2000/03 bear market: precipice bonds. These left capital exposed to losses only if the risk asset on which the option was written fell by more than a particular level. Often the risk in the structure was then leveraged, so that the capital eroded at twice the rate, in many cases, as the underlying. Viewed through the eyes of the counterparty, this structure represented the sale of catastrophe insurance, as it laid off its exposure to extreme loss in the underlying. Though extreme outcomes had a low chance of arising, the consequences were enormous. That meant the option cost was lower and the upside participation for the buyer looked much more enticing. It was also a smokescreen for increasing the cost taken out of the structure by the issuer, counterparty and distributors. This was clearly unsuitable for the customers who typically provided that catastrophe insurance to financial institutions, such as little old ladies popping into their local branch of their bank.)
I regard the distinction between capital and interest implicit in option-based structured products as being one that many customers would instinctively treat as specious. They are right: it is just money. The regulatory framework effectively reinforces a non-economic distinction between the components of money, as between capital and its rental cost. This is exactly the sort of semantics smart but conflicted agents will seek to exploit.
(The distinction between income and capital does have an economic dimension because of the different tax treatment of income and gains; the FSA chose to refer to this in its review but it misses the point that the tax treatment is part of the bet and cannot be treated as an independent test of the suitability of the bet.)
The structure is usually illiquid so the payoffs are specific (barring particular features or ‘bells and whistles’) to the level in, say, five years rather than at any stage within the five years. This is also a characteristic of most structured products that the FSA highlighted in its review, applying a narrow test of whether the customer could afford to tie up cash for the term of the structure. This will also be difficult to assess after the event if complaints are made.
Suitability of optionality
What might represent a logical, hindsight-free test of suitability of an option for an individual saver or investor? I think the answer is very limiting: a specific view at a specific time with a specific cost and a specific probability distribution for the payoffs.
In other words, it has to be a specific bet. But here is the key insight. If we assume that option prices are generally efficient, which for the regulator has to be the only safe assumption, it is a form of bet that becomes irrational if rolled over across serial structured products to become a permanent alternative either to cash or risk assets. It becomes irrational for the same reason that any form of permanent ‘portfolio insurance’ is, before the event, likely to be self-defeating, namely that the cost of the downside protection is likely to be equal to the cumulative payoff before costs. This follows because i) option prices are based on volatility and ii) risk premiums relative to cash are also a function of volatility.
Of course people value options, particularly when at no cost or very low cost. In such cases, they need not even have a view about the underlying. But if they (or their advisers) believe they have no good basis for arguing an option has been seriously mispriced, they implicitly need to base their bet on a specific view of the underlying.
In fact, for much of the period in which capital-protected structured products exploded in popularity, it was possible to argue that volatility in equities and option prices were both unusually low. These were probably circumstances associated with excessively lax money and credit growth which of course ended with the banking crisis (to which these same factors had contributed so much). For much of this period when volatility was low so was the interest rate given up, so the upside participation was also less. Taking both factors together, pricing was arguably a reasonable basis for some wealthy clients to add optionality to their cash deposits, particularly if they were high income tax payers, for as long as these conditions persisted.
I do not believe most products were sold on the basis of specific views of pricing inefficiency or of the underlying asset, except perhaps in private banks and wealth managers with sophisticated clients. Instead, I think they were mainly sold to risk-averse savers and investors as a permanently suitable alternative to deposit-based savings. Logically, these are the people to whom the non-economic distinction between capital and interest is most important, as they can least afford to lose the economic rent on their capital. This is not just because they are financially constrained but because the rent has to provide their inflation compensation.
However, it is also likely that they were promoted to many investors with high enough risk tolerance to have exposure to risky assets, particularly if they have long-term financial goals that are expressed in purchasing power terms, after inflation. For these investors, substituting unhedged exposures to real assets by hedged exposures with very low probable payoffs was unlikely to improve their welfare or utility. Comparisons of hedged and unhedged exposures are also sensitive to the difference between the two in terms of liquidity, as most structured products have in the past required giving up the ability to alter exposures with changing asset prices as few provided reasonably liquid secondary markets.
Advisers substituting risky exposures with hedges in conflict with the client’s ‘normal’ risk preferences were also singled out by the FSA but this looks an improbable target for enforcement. Advisers will probably argue that they realised at the time that the conditions that caused the low option costs were also ones that risked blowing up the economy, so the downside protection was a logical bet even if normally they would prefer unhedged exposure. Hindsight is a wonderful thing.
The FSA’s review appears to provide an incentive to customers holding illiquid structures which are currently unexpired but out of the money, or which previously expired out of the money or with very low returns, to challenge the advice given. Even without problems assessing advice after the event, this is a potentially horrendous burden for an already overworked complaints process.
I do not believe the FSA’s review provides customers, distributors or the FOS with a logical basis for assessing suitability except in extreme cases where a direct stipulation was breached and the breach was in genuine economic terms, not specious technical terms.
The obvious conclusion to draw is that the generic attributes of option-based structures, genuinely highly complex in terms of utility theory, needed to be addressed by the FSA when structured products became popular, not when the first counterparty happened to fail.
The FSA has also had years to decide that the sales process should have included estimates of the probability distribution of the payoffs, so that customers could see that most outcomes were bunched around zero or a small gain and that their ‘target’ or headline return had only a tiny chance of being achieved. Without it, issuers and distributors had an information advantage they could exploit.
Under the leadership of Lord Turner the FSA has had the humility to suggest that maybe product regulation is necessary and should not be ruled out. Was it asking too much to see some of that humility in this review?
Finally, the FSA should restate its position on structured products in the context of tests of independence when RDR takes effect in January 2013.