Firms have been queueing up to announce they will not charge clients for external securities research. But when MiFID II goes live, it will not be the end of the story. Fiona Rintoul reports.
The law of unintended consequences might have been invented for the second Markets in Financial Instruments Directive (MiFID II). With some investment banks suggesting fixed income research might be treated differently from equity research and a growing number of asset managers reneging on a previously held position that they would not take the cost of securities research on to their balance sheets, the consequences are only now becoming apparent.
“After January 3 [when MiFID II goes live] we will see what it means,” says Stefan Barkhausen, a spokesperson at Union Investment. “Even then it will be too early to tell the impact.”
The open questions are many. What is research? A morning newsletter on the markets that tells you stuff everyone knows already may not qualify if it has a price tag attached, suggests Barkhausen.
“It has to really add value,” he says. “Broader research available from diverse sources may not be interesting.”
There are other questions too.
“Are there potential tax implications from unbundling?” asks Anthony Kirby, a regulatory reform and risk management specialist at Ernst & Young. “Will the supply of research from the sell-side stay the same?”
On the latter question, a consensus is developing that supply will reduce in line with the decreasing demand that Barkhausen outlines. In a recent report on research unbundling, Oliver Wyman predicts a reduction of 10%–30% in research spending, though the reduction will not be uniform. In some cases, there will be little change; in others, research spending could be reduced by as much as 50%.
Some asset managers believe this will make a bloated market more efficient to the benefit of everyone – not least end investors.
“There is a lot of overcapacity with regard to sell-side research,” says Lars Dijkstra, CIO at Kempen Capital Management, which was quick out the traps in declaring it would pay for research. “When the buy-side starts to pay for this out of their own pockets, this overcapacity will be reduced and redeployed somewhere else. This will make the industry more efficient. End clients will also receive better value for money.”
But is it necessarily so? It may be premature to celebrate price savings.
“The price could go up,” says Kirby, “and different countries might apply different models.”
The truth is we just don’t know. In the meantime, asset managers are doing their best to adjust to the new rules, so far – in the main – by flipping over and saying that they will take the costs on to the balance sheet.
“The most pivotal announcement was when JP Morgan Asset Management said it would pay hard for research,” says Kirby. “That had an electric effect.”
‘Was it any good?’
Many in the industry feel that this is now the direction of travel. For one thing, paying for it yourself avoids the tedious explanations to institutional clients about which research was used and what the alternative – the research payment accounts (RPAs) introduced by MiFID II – implies. “Do you really want to spend fund managers’ time thinking why did I buy that and was it any good?” asks Richard Phillipson, principal at Accenture Investment Management Business Insight. “Already fund managers are not spending as much on alpha generation as they might like.”
There is also the danger that institutional clients might say no to the research bill. Some institutional clients, such as UK local authorities, are already looking to spend less with investment managers, notes Phillipson.
However, it is by no means a done deal. A flurry of announcements by larger firms may be masking a different reality in other parts of the industry. In an August survey by S&P-owned Crisil Global Research & Analytics, which contained a lot of smaller asset managers, 38% of the 92 managers surveyed said they planned to charge clients using RPAs, with 59% undecided despite the approaching deadline.
An earlier Financial Times survey of about 60 investment managers found half of them measured by assets under management (AuM) to be undecided, with 60% of those that had decided (again by assets) choosing to absorb the costs.
It is no surprise that smaller firms are less keen to absorb the costs. As Oliver Wyman points out, research is a cost that does not increase in direct proportion to AuM but has more to do with breadth of offering.
All the same, there is bound to be a momentum – unkind observers will call it ‘herd’ – effect.
“The industry is one that works on consensus and it would no doubt be more challenging to explain why some firms charged their clients for research whilst others absorbed the cost into their balance sheets,” says Gerard Walsh, head of business development, institutional brokerage, at Northern Trust Capital Markets.
“There are pros and cons to either model but it now seems larger managers will take costs in-house, which is a simpler budgetary and allocation process in the round.”
The German experience
The German market provides an instructive example of how the situation is evolving, at least among the larger companies. All four of the major German players – Allianz Global Investors, Deka, Deutsche Bank and Union Investment – initially said they would pass the cost of research on to clients when MiFID II came in. Since August, all but Deka have declared in the other direction.
At Union Investment, one of the rationales for the change of direction was to cut costs. It’s hard to dispute Barkhausen’s argument that transaction costs should be lower than before without the research. Union Investment also has a clear strategy for minimising research costs. It will use ‘really good’ external research and will expand internally.
“That was ongoing anyway,” says Barkhausen. “Every fund manager at Union Investment has a research task as well. They have to be an industry specialist.”
But the eventual cost of the external research the firm does continue to consume remains opaque. This a commodity that has not been priced before.
“This approach [unbundling broker commissions] is simple and probably very pragmatic in the circumstances, but clearly opens up opportunities for broker research to be miscalculated,” says Cosmo Wisniewski, director at Citisoft.
“In the worst cases, the clients could be impacted financially as their execution-only commissions may not be quite what they appear.”
There is also the question of the mechanics of charging for the research. Daniel Carpenter, director of Meritsoft, which helps brokers track trade executions and research commissions, estimates that there will be a 100%-400% increase in the number of agreements that need to be put in place.
“Houses on the sell-side have to industrialise current processes,” he says. “It will go on well into next year.”
Certainly, buy-side firms will want to scrutinise the cost of research more closely if they are paying for it directly. “There’s a step change when CFOs become more interested in the budget than they used to be,” says Phillipson.
“It becomes a matter of procurement like anything else.”
And well they might scrutinise these costs. Oliver Wyman’s estimates suggest that research costs make up on average only one to three basis points of the total active management charges of around 60 basis points that are ultimately borne by end investors. However, absorbing these charges could add 2%–4% to operating costs for asset managers – equivalent to a 4%–7% profit reduction.
It doesn’t take a genius to work out that the screws are going to be turned on investment research. Against this background, a number of platforms have emerged that aim to create an efficient market in research.
One such is Smartkarma, which covers the Asian markets. Co-founder Jon Foster describes it as the Spotify of research. Operating in a post-MiFID environment, platforms such a Smartkarma will democratise the provision of research, he believes.
“I don’t think people will buy less research, but we will see an end to massively inefficient research production and distribution techniques.”
Crucially for those smaller firms concerned about research costs, Smartkarma charges on a per-user basis. There may, however, be a deeper problem with sell-side research than just oversupply or a charging mechanism that disadvantages smaller firms. Some suggest that the sell-side is supplying the wrong kind of research.
“Currently, the sell-side industry in general produces research with an horizon of around six to 12 months,” says Dijkstra.
“This horizon matches less and less with the average horizon of its client base, which has bifurcated into very short and long to very long. At both ends, the buy-side has been investing in its own research capacity.”
As the MiFID II deadline approaches, many questions remain – not least if global firms will handle non-European accounts MiFID-style. Answers are in shorter supply, and some, such as Citisoft’s Wisniewski, believe too much change in the industry is being driven by burdensome regulation.
“This one will roll on and on, and we expect a rumble that continues for 18-24 months at least,” he says.
®2017 funds europe