PsyFi Search

Thursday 17 September 2009

Disclosure Won't Stop A Conflicted Advisor

Monetary Conflicts

Nowhere are conflicts of interest quite so common as in the financial industry. Accountancy firms want to provide consulting services, credit rating agency employees want to work for security analysis firms, independent security analysts need to ensure their recommendations don’t lose their employers important mandates, financial advisors want to get commissions for recommending products ... the list is almost endless.

The popular answer to these problems is almost always the same – disclosure of the conflict of interest so that the purchaser is fully aware of the potential problem. Unfortunately the reason that disclosure is so popular is that it doesn’t work. Even worse, it may actually make the problem worse.

Corruption’s Not The Problem

Let’s start this sad catalogue of problems with an important caveat. In almost all instances the problems caused by conflicts of interest between personal interests and professional ones are not cases of corruption. Of course there are some advisors who are blatantly dishonest and who deliberately oversell their products but these are relatively few and far between and besides aren’t the topic of our investigation. No, our bunch of conflicted advisors are driven by internal demons that are far more difficult to pin down. These are unconscious side-effects of various psychological biases rather than deliberate acts of criminal intent.

In the end most corrupt advisors get rumbled – although as Bernie Madoff has shown, this may take an awfully long time. The problem with the honest ones is that they’re still chiselling us yet would be able to fool a lie detector or, worse, their wives, without the slightest difficulty. Meanwhile the methods used to protect us from these practices may actually make things worse.

Disclosure or Regulation?

The most popular remedy to conflicts of interest is disclosure. The idea is that if the advisors, of whatever hue, declare their conflicts then we, the purchasers, can make a clear sighted call about whether the advice we’re being given is in our best interests. However, we can almost immediately identify two problems with this. Firstly, if we were really able to make such decisions then we wouldn’t need to consult an expert anyway. Secondly, even if we do adjust our behaviour to take account of the conflict we have no guarantee that the advisor won’t adjust theirs to take account of the fact we know about the conflict.

And if you didn’t get that don’t worry. We’ll get back to it soon enough.

The alternative to disclosure is often regulation, making the disclosed behaviour illegal. Mostly the affected parties prefer disclosure since this avoids them having to give up whatever they’re disclosing and also means that all parties don’t have to adjust to a difficult new reality. However, the idea behind disclosure is actually pretty dubious: it shifts the responsibility for making informed decisions from the (supposedly) knowledgeable advisor to the (usually) uninformed client.

The Dirt on Coming Clean


Cain, Loewenstein and Moore investigated this whole area in The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest and anyone really interested in this topic should peruse the source material as it’s impossible to do it justice in a short article. It's a very fine piece of research.

They took the basic idea of disclosure into a laboratory setting and set about developing an experiment to investigate its effects on advisors and clients. By comparing advisors' own personal estimates with the advice they actually gave under different conflicted situations they found that:

1. If advisors don’t have any conflicts of interest their advice is unbiased;
2. If advisors do have conflicts of interest but don’t disclose them then their advice is biased;
3. If advisors do have conflicts of interest and do reveal them then their advice is even more biased.

Trying to disentangle the various threads involved in this is tricky but two possible explanations have been proposed. The first is simply that the disclosure of conflicts of interest leads advisors to assume that clients will discount their advice, so they increase the bias in their advice to compensate. The second is that the disclosure frees the advisor from guilty feelings about violating professional norms and this so-called “moral licensing” leads to more bias.

There was another problem, though. Although clients discounted their advisors’ opinions when the conflict was disclosed this was insufficient to offset the bias. Which led to even less accurate decisions than when the conflict was undisclosed.

Clients Are Also Conflicted

So this is a two-way street: clients are also affected by advisor disclosure, only not necessarily in the way you might expect. Other research has found that disclosure doesn’t necessarily deter clients from accepting advisors’ advice and may actually have the opposite effect. Two underlying biases seem to be implicated in this. Firstly we have a tendency to delegate difficult decisions to experts – so, for instance, we generally accept the opinions of our doctors without ever asking for a second opinion (a mistake, by the way). So the fact that we’re being told that the expert has a conflict of interest often doesn’t make much difference – the expert is the one in charge.

Secondly, the disclosure inadvertently seems to trigger a response that works along the lines of “if they’re told me about this conflict then that shows that they’re honest and if they’re honest then I should trust them more”. The fact that the honesty has been forced upon the advisor by a legal requirement to make the disclosure is irrelevant, it seems. It’s the disclosure that triggers the response.

Expert Evaluation of Expert Disclosure

Possibly the biggest problem, however, is that a disclosure of a conflict of interest by an expert often requires an expert to evaluate it. If you go to a doctor about a difficult medical problem and the physician tells you that you need an expensive drug but that she’s being paid by the pharmaceutical industry to plug it how do you respond? Does this disclosure help you make the decision or not?

Well, not, as it happens.

If you can’t trust the expert giving you advice then possibly your only option is to get another expert opinion. However, the evidence from medical research is that people don’t bother. This is possibly more worrying in finance than medicine. As Bill Bernstein has said:
"The depressing fact of the matter is that the federal and state governments do not protect investors in the same way that they do the clients of other professions. Amazingly, stockbrokers are not considered fiduciaries at all, as are practitioners of all other learned professions. This state of affairs is, in some respect, a historical accident.

All the professions I've mentioned, except finance, have long since recognized that regulation of minimal standards of training and practice is a necessity. It happened, for example, to the medical profession a century ago with the publication of the Flexner Report. Bluntly put, there's no chance that your doctor, dentist, or attorney is a high-school dropout. Your stockbroker, however, just might be."
After all, what is a financial expert anyway?

You Get What You Pay For – One Way Or Another

Disclosure’s a lousy way of fixing the kind of problems caused by conflicts of interest. It seems that business models that rely on promoting such conflicts are fundamentally flawed because they’re bound to trigger this type of behavioural confusion. In fact, wherever you find a situation where disclosure is mandated you should automatically assume that you're at risk from these types of unconcious bias. As my grandmother used to say: you get what you pay for. Even if you don’t know you’re doing so.

The research says, that when it comes to advice, especially in financial matters, it’s always safest to pay directly rather than indirectly. Alternatively, get a second opinion.


Related Articles: You Can't Trust The Experts With Your Investments, Bulletin Boards Are Bad For Your Wealth, Survivorship Bias In Magical Mutual Funds

8 comments:

  1. There is a very easy way of knowing if your stockbroker is a high-school drop out, it is called asking a question. I was with you up until that point. How do I know that the medical certificate hanging on the Dr's office isn't a 3-day pharma paid beach resort golf paying long weekend with a couple of seminars in the morning?

    Your implicit assumption is to avoid trusting a conflicted advisor but blind trust of regulation (wasnt Bernie Madoff highly regulated?) is the answer? I believe everyone should do what is best for them. And expect that everyone else is doing the same. And take even paid advice with a grain of salt and use your common sense. And if an advisors incentives is not aligned with yours, they are probably a salesman more so than an advisor anyway.

    Sorry for the rant, I like the blog a lot and recommended to many people in finance.

    ReplyDelete
  2. I don't think that's a rant: seems perfectly reasonable comment. Regulation clearly isn't the whole answer but enforeable minimum standards, the threat of being struck off and losing your livelihood and an industry funded compensation scheme would be a start.

    Blind trust in regulation is always stupid but the reality is that this is what most people do with advisors of all kinds. All regulation can do is reduce the problem. The alternative is all problems being solved by the free market and the evidence, at the personal level, is that it can't. So I think we need a balance and that the balance is currently wrong because the alternative - disclosure - makes the problem worse, not better.

    Glad you like the blog: all opinions (if politely put) are welcome.

    ReplyDelete
  3. I love the post. It makes a very important point that is rarely made.

    I don't like Bernstein's example about the high-school drop-out. What matters is whether the advice is good or not, not how many years the person giving the advice went to school. Bernstein offers some great advice. Berntein also offers some horrible advice. The trick is separating the good stuff from the bad stuff.

    Bernstein is highly intelligent. That may contribute to the good part of what makes some of his advice so good. It also contributes to the bad part of what makes some of his advice so bad. The single biggest problem with investing advice is that so much of it is the product of rationalization. What makes someone a great rationalizer? Intellect. Intellect is often a negative in this field.

    I'm not saying to look for dumb people to give you your investing advice. My take is that you need to drill down and make sure that the advice being offered makes sense. The behavioral problem is that most investors decide in advance what they want to hear and then go searching for an "expert" to tell them what they want to hear. "Experts" with impressive educations are more persuasive in telling us stuff that we want to hear that will probably cause us to lose lots of money.

    Rob

    ReplyDelete
  4. I'd love to see if these results generalize to physicians giving pharmaceutical industry sponsored talks on treatments for condition X.

    Forced disclosure is common in talks. Does this turn doctors into biased promoters in the same way as demonstrated with financial advisers?

    ReplyDelete
  5. I think the question of intellect is an important point. On one hand over-analysis and the use of too much information has been shown to yield worse results in a number of fields. More information actually makes it harder to see the wood for the trees, it seems.

    On the other hand I think we need to be careful about our own intellects. It's all very well for us to sit around and discuss PE/10 and q and so on but the vast majority of people have no idea what we're on about and not much ability to change that state.

    Arguing that the average Joe or Joelyne should learn about this in order to protect themselves may simply expose them to the vast range of psy biases that get discussed here for no extra benefit. There is a place for a know-nothing, good enough approach and otherwise aligning advisors and investors interests should be a priority.

    On physicians I don't know any details around sponsored talks but other research seems to show that they are also conflicted by pharma gifts and so on, so it wouldn't be surprising if this is a problem.

    ReplyDelete
  6. It's all very well for us to sit around and discuss PE/10 and q and so on but the vast majority of people have no idea what we're on about and not much ability to change that state.

    We disagree strongly on this one, Timarr.

    I have spoken with thousands of middle-class investors and a good number of big-name "experts" re the valuations question. Everywhere it comes up the reaction I get is the same. The idea of taking valuations into consideration makes perfect sense to most middle-class investors. Middle-class people take prices into account when buying everything they buy (except stocks). So the idea of investing rationally is a natural for them.

    There's one big problem. Most middle-class investors believe that the "experts" in the field must have some idea what they are talking about and the "experts" have been telling us non-top for 30 years now that there is no need for investors to adjust their stock allocations when prices go to insanely dangerous levels. Millions of investors were bullied into investing recklessly with their retirement money and we have experienced the greatest loss of middle-class wealth in the history of the United States as a result.

    I think it is fair to say that it is not pure coincidence that The Stock-Selling Industry obtains a huge financial payoff for promoting Passive Investing. The downside is that this "strategy" always ultimately causes an economic crisis when the investors following it are wiped out. This must stop.

    "Know Nothing" investing advice doesn't cut it. The promoters of Passive Investing say that it is based on academic research. The reality is that the academic research has been showing for 28 years now that valuations affect long-term returns (and that the chances that Passive Investing could ever work in the real world are thus precisely zero). When we tell people that we are reporting what the historical data says, we take on an obligation to report what the historical data says accurately.

    Rob

    ReplyDelete
  7. I hope the FSA takes this research into account when it finalises the RDR. It's recomendation of having a "Sales Adviser" seems to be the worst of all worlds.

    Moreover, advice need not be binary, as in good or bad. There are shades of grey, just as there in hotels ranging from Fawlty Towers to the Dorchester. But how you assess 2 or 3 star advice, and how much you should pay for that, is a tricky issue.

    ReplyDelete
  8. Hi Rob

    S'Ok to disagree: only way we ever advance our thinking :)

    Obviously I can’t comment on the investors that you’ve talked to personally although I certainly don’t doubt their interest. However, there’s the ever-present danger of extrapolating from a self-selecting set of people – those who seek advice and are genuinely interested ought to be the least of the problems. Unfortunately the evidence also suggests that even interested people find it hard to overcome their biases if they’re actively involved in making investment decisions.

    I’d agree that most people are comfortable taking price into account but I would dispute that they’re very good doing the same with value. There’s plenty of evidence that people use price as a short-cut for “quality” – it’s quite common for stores to increase the price of goods to shift them when they’re not selling. As ever price is what you pay, value’s what you get.

    Personally I don’t think purely passive investing in a single asset class is a particularly good idea but I also don’t think any approach requiring active investment decisions to be made by all individual investors is sensible either. Some kind of quasi-mechanical-passive approach across multiple asset classes with automated rebalancing would seem to be indicated: although that sounds like a peculiar echo of the efficient frontier to be honest ...

    ReplyDelete