Field Notes Is AIMR too "sell-side"?
IN PRINT ARCHIVE CIR Winter 1999
|Is AIMR too 'Sell-Side'?|
|by Keith Ambachtsheer|
Have you ever sat on a committee or board with a pain-in-the-butt member who raises the same issue meeting after meeting? If you have, you have our sympathy! This writer increasingly became one of those "oh no, here he goes again" people during his six year tenure on AIMR's Council of Education and Research. Our issue was that in structuring its education and research programs, AIMR not lose track of who the ultimate clients of the providers of investment management and research services are.
This descent to single-issuehood was at least partially triggered by an exchange between the Council Chair and this now Ex-Council Member a few years back. The cited exchange below is not necessarily accurate verbatim. However, it does capture its essence:
Ex-Council Member: "We don't seem to spend a lot of time discussing 'buy-side' issues on the Council."
Council Chair: "We do. The perspectives of institu-tional investors get as much time and attention at the Council as those of Wall Street."
Ex-Council Member: "That is true. However, institutional investors are just as much part of the 'sell-side' as broker-dealers. The ultimate clients of investment management and research providers are pension plan members and sponsors, endowment fund beneficiaries, mutual fund holders, etc. Directly or indirectly, it is out of their assets that all fees, commissions, and spreads are paid. An important question is whether these ultimate 'buy-siders' receive a proper balance between what they pay, and the considerations they get in return."
We will set out possible implications of this observation for AIMR's education and research programs, and for the future of AIMR itself. Then we will get off our soapbox...for a while, anyway!
|Understanding the Market|
A fundamental tenet of micro-economic theory is that the interplay of buyers and sellers in free markets matches the demand for and supply of goods and services in the 'right' amounts, at the 'right' prices. Consequently, everyone is better off. However, there is an important (but often overlooked) caveat to the tenet. It is that the buyers and sellers have equal knowledge about the nature and quality of the good or service being bought and sold. When this assumption does not hold, all bets are off.
The economist Ackerlof eloquently demonstrated this point 25 years ago in his celebrated article "The Market For Lemons." Why is there an immediate price drop once a new car becomes a used car, he asked. The answer is that in the market for used cars, the sellers have more specific product knowledge than the buyers. In other words, the sellers know whether or not their cars are lemons, but the buyers do not. Sensible buyers of used cars demand a price discount to compensate them for this 'informational asymmetry' favouring the sellers.
What if buyers underestimate the probability they may be buying a lemon? Then they will pay too much for too little. We believe this latter situation continues to reflect the reality of the market for investment management and research services. Specifically, the 'buy-side' has histori-cally overestimated the value active investment management services can deliver, on average. Thus they too have paid too much for too little, on average.
Research on the organizational performance of defined benefit (DB) and 401(k) pension funds supports this proposition. For example, in a $1.5 trillion pool of 124 U.S., Canadian, and Dutch DB pension funds, the median fund underperformed its policy benchmark by about 50 bps. per annum over the last five years.
|Levelling the Playing Field|
Is there a way to level the informational playing field between the buyers and sellers of investment management and research services? Absolutely. Effective 'buy-side' governance is the answer. In a pension fund context, this means the assembly of a group of governing fiduciaries willing and able to create value for the pension fund stakeholders.
The governors need to have a clear idea of who the stakeholders are, and who is underwriting what kinds of risks. They need to understand the difference between governing an organization, and managing it. The governing they do, the managing they delegate to the organization's CEO or equivalent.
The CEO proposes a business plan. When approved by the governing fiduciaries, it becomes the roadmap which guides the organization. The roadmap is shaped by an organizational vision and a realistic assessment of where the organization's comparative advantages lie. It leads to an internal staffing plan, and to plans for the identification of, and optimal contracting strategies with outside providers of investment management and research services.
State-of-the-art information technology monitors risk exposures, and creates feedback loops consistent with the architecture of the organization. The governing fiduciaries get the total fund oversight information they need. The managing fiduciaries get the business unit information they need. The operating fiduciaries (internal and/or external) get the individual portfolio level information they need.
All this is not just a vacuous recitation of organization design principles. Our research shows that pension funds which build strong organizations really are more likely to enjoy superior investment results. The reverse is also true. Pension funds which do not explicitly organize to create value for stakeholders are more likely to actually suffer poor investment results. They tend to pay too much for too little.
Specifically, we have found that well-governed and managed DB pension fund organizations outperform poorly governed and managed ones by 90 bps. per annum, all else equal. This research is documented in an article in AIMR's own Financial Analysts Journal ("Improving Pension Fund Performance," Keith Ambachtsheer, Ronald Capelle, Tom Scheibelhut, Financial Analysts Journal, November/December 1998).
|From Pension Funds to Mutual Funds|
If pension fund stakeholders fall prey to poor investment results because their governing fiduciaries have not built strong 'buy-side' organizations, what about mutual fund unit-holders? The poor performance potential here is even greater. Why? Because the directors of mutual funds do not have the option to create their own strong 'buy-side' organizations.
By design, they must use the investment management and research services of the 'sell-side' advisory organization which created and markets the funds the directors are supposed to govern. The management of these advisory organizations typically have mixed allegiances. On the one hand, they really would like to create value for their unit- holders. On the other, they are expected to create value for the owners of the advisory organization even more. Unfortunately, the financial interests of the former and the latter do not always coincide.
We are not alone in expressing concern about the potential impact on mutual fund unit-holders of these asymmetric arrangements. In a recent speech on mutual fund governance, the SEC's Chairman Arthur Levitt proposed to strengthen the independence of mutual fund directors in a number of ways. In Canada, the OSC has also commissioned a study on the advisability of creating independent investment fund governance mechanisms. However, we believe that independence by itself will not be enough. Mutual fund directors must also be able to hold the fund advisor accountable for pre-defined results, pay fees based on the results actually achieved, and terminate the advisory contract in organizationally dysfunctional situations. There is a long way to go before the typical mutual fund unit-holder is able to invest on an informationally level playing field.
What does all this have to do with AIMR's education and research programs? Here are some of the connections:
How much do AIMR members care about delivering 'value for dollars' to their ultimate 'buy-side' customers? If they do care, what does that imply for how investment performance should be measured and communicated from a 'buy-side' perspective? What are the education and research implications?
Should an explicit AIMR goal be to reduce the informational asymmetry between the sellers and buyers of investment management and research services? If the answer is 'yes', what strategies would be most effective? What are the education and research implications?
Is AIMR interested in the effective governance and management of such fiduciary vehicles as pension, endowment, foundation, and mutual funds? If the answer is 'yes', what are the education and research implications?
Below we explore just one of these issues. Specifically, we assess the efficacy of the much-touted AIMR investment performance presentation standards as a useful 'buy-side' information tool.
Are the AIMR PPS Standards a useful 'buy-side' tool? We would answer "yes and no." On the 'yes' side, the Standards represent a giant step forward in creating some order in the previously chaotic world of 'sell-side' investment performance reporting. Even the dullest 'buy-siders' were aware that in a world where all presenting investment firms are able to demonstrate first-quartile results, something was desperately wrong. The PPS Standards undoubtedly make the slights-of-hand which conjured up these 'everybody is in the first quartile' results far more difficult to perform.
On the 'no' side, the PPS Standards perpetuate and even strengthen the premise that there is a useful positive correlation between past investment performance and future investment performance, when properly measured. Unfortunately, the bulk of available empirical evidence does not support this proposition. This raises two further questions for an AIMR truly concerned with the financial welfare of its ultimate 'buy-side' constituents. First, why is it that there appears to be no useful positive correlation between past and future investment performance? Second, if not past performance, are there other useful predictors of good future investment performance?
The questions we pose above lead education and research in a potentially uncomfortable direction for an organization whose 36,000 members in over 80 countries (including Canada) largely make their living on the 'sell-side' of the market for investment management and research services as we have defined it. For example, a plausible explanation for the past-future 'no correlation' performance findings is that most 'sell-side' active management investment firms produce long term excess returns of zero, blurred by a short term noise band which is sometimes positive, and sometimes negative. In Ackerlof's terms, these firms are lemons.
So how does the ultimate 'buy-side' identify investment firms with future excess net returns which are more likely to be positive than negative? Clearly, the starting point is to develop strong priors about the key characteristics of non-lemon investment firms. This is not the time or place to pursue such a quest. The point here is that given the informational asymmetry reality of this industry, the direction professional education and research takes needs to be guided by a view on how the respective interests of the buyers and sellers of investment management and research services are to be served. Depending on the balance of respective interests chosen, the direction AIMR education and research takes can vary considerably.
Which gets us back to the title: is AIMR too 'sell-side'? Note we say "too." AIMR must be a 'sell-side' organization, because the vast majority of its membership is employed by the 'sell-side' of the industry as we have defined it. That is not our issue. Our issue is that AIMR must not lose sight of who its ultimate customers are, and how their financial interests are best served. It would do so at its peril.
Keith Ambachtsheer is president of KPA Advisory Services Ltd.