Regulation means that fund managers are submitting large quantities of data to regulators. Though regulators have yet to do anything with it, hedge fund operational advisors Convergence LLC are proving that private sector firms can find plenty of valuable uses for the information which is available from just one form: the Form ADV submitted to the SEC by every manager of third party money in the United States.
The original ambition of the founders of Convergence LLC was to help hedge fund managers cut their operational costs. That is still their aim, but they have over the last 12 months evolved a new way of doing it. A business based on consulting and advice [see “How to benchmark and then cut your operational costs” at http://cooconnect.com/online-magazine/how-benchmark-and-then-cut-your-operational-costs-0) is now a data business. It is a transformation many much larger businesses would dearly love to accomplish. Convergence has managed it by scripting software to rummage through the vast quantities of structured and unstructured data submitted by fund managers to the Securities and Exchange Commission (SEC) once a year via the Form ADV that all registered investment advisers must complete.
The Convergence detection machinery has even waded through the brochures and supplemental schedules managers are obliged to attach to their Form ADV submissions, because these contain crucial information about investment strategies and fee structures. To these regulatory sources, the principals at the firm have added another publicly available source of information: the presentations investment managers have given to public funds they are pursuing for allocations. They have also instructed their machines to trawl through the web sites of every manager who has submitted Form ADV. In a number of cases, where managers have become clients of Convergence, the firm has also taken the opportunity to acquire further information directly. Lastly, the machines search the Internet constantly for any other information relevant to the fund managers in the Convergence database.
According to John Phinney, co-managing partner at Convergence, the software-led searches produce something fit for immediate transmission to the database in nine instances out of ten. Once the analysts and technologists have dealt with the exceptions - of the ten people that work at Convergence, six are technologists and two analysts - the data is aggregated, de-duplicated, and pared of redundancies. It is then married to other information derived from another SEC document (Form 13F) that managers must complete once a year, and to a variety of data sources proprietary to Convergence.
The result is an exceptionally rich and accessible database of information about the post-trade infrastructure of alternative investment managers, which is helping Convergence build relationships with managers in unexpected ways. “We have spent over 25 man-years in the last 18 months developing the data and the technology to extract, understand and access the data,” says John Phinney. “We have catalogued and mapped 8,000 web sites, and compared what is said there with what the same firm says in their Form ADV. There are plenty of inconsistencies, where we actually call the manager, and ask them to qualify what is said. They are very grateful, because the SEC is all over this, and if your web site and your Form ADV are not aligned, you are going to have a problem. Behind those insights lies a lot of work. We have spent several million dollars developing very sophisticated technology capable of receiving, cleaning and parsing a terabyte of data every day.”
That said, Phinney insists that the firm is offering a service rather than data or data processing technology. One sign of that is the updates mailed to users as new information comes in, even though Form ADV itself is not updated more than once a year. Convergence publishes changes affecting managers, whether gains or losses of assets under management or a new fund launch or the appointment of a prime broker, on its web site. “We are publishing hundreds of original pieces of news on our web site every day,” says Phinney. “For us, an original piece of news is a change in the 15 million data points that we examine day in and day out. Perhaps a firm launched a new fund, or extended the type of manager they are from private equity only to include a hedge fund as well. We have identified 35 different changes that our clients say are unique enough to be notified to them.”
To generate that volume of news, Convergence compares news announcements with the contents of Form ADV, allowing the firm to create news bulletins on new fund launches, increases and decreases in regulatory assets under management (RAuM) at particular managers, new mandates won by fund administrators, fresh appointments as auditors, increases and decreases in the number of employees at management companies, and even changes in the regulators to which managers must report.
As Form ADV is updated in the first quarter of the year, the Form itself becomes a source of news, as the new submissions are compared automatically with the old. All of these snippets of information are useful to someone. “To the head of sales at a fund administrator, following 100 prospects, the news that one of those prospects has launched a new fund and has yet to appoint an administrator, is extremely valuable,” explains Phinney. “He or she will be one of the first to know it is going on. We are trying to give our clients first mover advantage through our information service.”
By its nature, the data is of interest to the middle and back office, rather than the front office. This is not unimportant, since it is a primary differentiator from the information services furnished by the likes of Barclay Hedge, Preqin and Audit Analytics, all of which focus on performance, assets under management, fees and other front office issues. Another differentiator is the exclusive focus of the Convergence data on alternative funds, and hedge and private equity funds in particular. Importantly, the Convergence data is also independent: unlike front office information services, the data is not derived from managers themselves. Even the consultancies which do collect middle and back office data, such as Cambridge Associates, tend to derive it from their clients only, which limits its use as the foundation of a wide spectrum of benchmarking services.
“This service is not designed for portfolio managers,” explains Phinney. “It is not about performance, or risk and reward attribution. It is designed for middle and back office professionals. It is about operational infrastructure. It tells you everything you want to know about a manager in terms of infrastructure. It helps COOs justify, support and track the decisions they make. Users of our service want to influence operational risk, headcount and back and middle office practices in a way that brings them closer to the mean. The mean may not be best practice, but it is a good place to be if you are raising and managing money. Getting managers closer to the mean is our core focus.”
This work - inching managers whose operational practices make them outliers back towards the middle of their peer group - draws Convergence into the custom research and consulting projects that the firm was founded to handle. The founders always expected, for example, to advise managers with over-staffed back offices on how to reduce numbers and costs. But John Phinney says the firm would now much rather partner with third parties to do that kind of work, because its real ambition has changed: it is to build a data business with reliable and repeatable revenues, rather than using data to support a consulting business that is by its nature hard to scale.
The pricing model reflects this change of focus. Instead of billable hours, it is subscription-based. Convergence offers managers an annual fee based on a sliding scale geared to the number of users, plus an annual technology maintenance fee. The price starts at $25,000 per annum for a minimum of 25 people, rising to a maximum of $50,000 a year for 50 or more users. These sums purchase access to the database and its search and selection tools only, without any additional services. Although users have to download the data into spreadsheets and develop their own uses for it, the basic service appeals to firms – such as the Big Four consultancies, auditors, recruitment agencies and the largest fund management houses - which employ data analysts of their own.
For firms which lack resources of that kind, Convergence has developed a series of eleven additional services (see the table below) which draw on the database, but which are sold separately. Prices for these services start at $7,500 per year per service. So far, the portfolios purchased have yet to exceed $25,000 from any one client. However, the products are less expensive only on an individual basis. A buyer purchasing all nine would pay $82,500 a year, plus another $25,000 for unlimited access to the database, or $107,500 in all. “We would tell someone who was thinking of buying all the products and the database to think twice about that,” says Phinney. “We would advise them to buy the database, and offer to work with them to show them how to create the products they need. We are not interested in gouging people. We want to give people what they need. If you buy the database, you can be self-sufficient in at least six of the eleven products, if you are savvy enough and prepared to invest the time in it.”
The eleven products rely on filtering of 2,000 “points of interest” that Convergence has identified in the data. Manager Benchmarking, for example, allows managers launching new funds or adding new strategies to estimate and benchmark the front and back office headcount the new fund is likely to require at different RAuM figures. Client Benchmarking Analysis, on the other hand, allows service providers to measure whether their clients have too many employees by comparison with their peers. “The service provider who finds that it has disproportionately more of the less efficient managers than its peers needs to ask itself if it is doing enough to help its clients,” says John Phinney. “That includes asking itself whether it is investing enough in its services to help them become more efficient. It is a great tool for service providers to understand better their book of business, and how the clients that make up that book of business rank in terms of efficiency against clients serviced by competitors.”
Another service, Regulatory History Analysis, looks at regulatory filings by managers dating back as far as 2002. Users can check which managers running which strategies have attracted the attention of which regulators in any one of dozens of jurisdictions, and document both the exact violation and any settlement reached. A fourth service, C-suite Contacts, draws on the disclosures in Form ADV about the senior people who work at fund management houses submitting the form. A related service, the Talent Acquisition Manager, focuses on the four top level C-suite roles: CEO, CFO, COO and CCO. It is proving predictably popular with recruitment consultants because it can quickly identify, for example, every CCO working at a distressed debt manager. On top of that, it can filter them for experience of different regulatory regimes and auditing firms, gender, location and a variety of other characteristics, including diversity. “For recruitment consultants, it is a convenient way of generating a slate of candidates for a particular job, especially since the list comes with full contact details,” says Phinney.
A sixth service allows users to calculate the market share of the administrators, auditors, law firms, prime brokers and custodians that service fund managers. Market Share Analyzer, as it is called, addresses directly a longstanding information gap in the fund management industry: the lack of reliable information about the market share of different service providers. Most market share estimates are based on information collected from the service providers themselves, which are not always reliable and lack granularity. “For us, market share means a lot of things,” explainsPhinney. “It can be calculated by investment strategy, fund type, manager type, size, or geography. We can cut the market share data in a lot of different ways. So we can be very granular about what a service provider is good at. ” A fund administrator, for example, can loom large in an overall league table, but have a minuscule share of a market segment in which it wants to build its book of business. Likewise, fund managers looking to appoint service providers with experience of particular strategies and geographies (and a sizeable market share to ensure continuing commitment) can use Market Share Analyzer to identify the right firms.
One benchmarking service could scarcely be more topical: fund expense practices. The Convergence software has now crawled through the complete texts of 8,000 fund management brochures in search of disclosures about how expenses incurred by managers are allocated between the management company and the fund. The Fund Expense Practices Analyzer Service, as it is called, aims to help managers and investors understand how managers disclose and allocate expenses between the management company and the fund. “We have identified 2,000 expense `terms,’ reduced them to 113 expense categories, and divided them between 53 major expense groups,” explains Phinney. “Legal expenses, for example, are a major expense group, but it has 15 or 20 categories, such as side-letter expenses and litigation support, and within those categories dozens of terms are used, depending on how they describe the expense. Everyone describes legal expenses differently, for example.”
There is obviously no point in comparing what an equity long/short manager does with what a distressed debt manager does, since their expense lines are so different, let alone the business model. By aggregating and allocating disclosure practices between funds by size and strategy, Convergence is able to build size- and strategy-specific benchmarking tools, which enable managers to assess their own practices against best practice in fund of similar size or those pursuing the same strategy. Importantly, the tools are enhanced by a statistical analysis to measure frequency, so managers can gauge common as well as best practices in expense disclosure and allocation. After all, any manager who discovers that a majority of its expense practices are unusual has learned something of immense value when competing for investment allocations.
Phinney cites a client that has not only changed its disclosure and allocation practices after learning that they were an outlier in the Convergence data set, but tightened its language to explain to investors that the “accounting” costs they meet represent the costs of the third party fund administrators and not something else. “Why use 35 words to describe something that can be described in two?” asks Phinney. “Often, managers need to do something as simple as tightening or enhancing the language of their disclosures, or give additional information in their disclosures, to bring themselves into line with best practice. We can in effect certify to an investor that a manager follows best practice in expense disclosure and allocation.”
So it is odd to learn that Convergence has not yet opened its services to investors as well as managers and third party advisers. Phinney explains that the firm felt it was more important to convince managers of the value of the services first. “We have not approached investors so far, and for one simple reason,” he says. “We believe that this industry has a way of educating itself about this issue thoroughly, without running the risk of investors over-reacting to information they have not had time to understand fully. We know that investors will eventually get it. By talking to managers, and persuading them to make use of our tools, we are helping them to prepare them for the call from the investor when it comes.”
Phinney cites the example of a $500 million credit fund that wanted to outsource its compliance function, but ran into objections from an existing investor that claimed no other manager outsourced compliance. A search of the Convergence database identified 400 credit managers that outsourced compliance. “By the way, the investor did not care – they just did not like the practice,” recalls Phinney. “But we did debunk the myth that nobody does it.” Similarly, Convergence was able to help a manager being sued for excessive use of leverage, by using its database to prove that the degree of leverage employed by the fund was far from unusual. “It is all in Form ADV,” says Phinney. “You just have to know how to piece it together.”
Undermining unsubstantiated beliefs and assertions by making the contents of Form ADV accessible certainly has the potential, over time, to change the intrinsically antagonistic relationship between managers and their investors. “The goal of Convergence, as our name suggests, is to bring the two sides of the industry together to establish common approaches to the same issue, and common ways of measuring performance against them,” explains Phinney. “That leads to a less confrontational and more lasting solution. Naturally, we were concerned that we would be perceived by managers as the enemy. By choosing to develop the information, socialise it, and then distribute it, we have avoided that risk. It has been a successful model, so far.”
There are reasons other than prompting investors to ask awkward questions that fund managers might be suspicious of what Convergence is doing. No COO is pleased to discover that he or she is running a relatively inefficient operation. Phinney is aware of the risk of embarrassing potential clients, but insists attitudes are becoming more positive. “Managers are really warming to what we do,” he says. “If we put this information in front of COOs and CFOs and CCOs, and they choose to ignore it, shame on them. It is true that, at the moment, we sometimes have three or four conversations with a manager before the light goes on. Hedge fund managers are among the smartest people in the world, and they eventually get it. The notion that we have this massive database, which is fully transparent, and enriched with the perspective of industry veterans, is getting through. When we talk about operational risk, and complexity, and especially about the risk of regulatory penalties and sanctions, we are talking about things any middle and back office practitioner is concerned about and would agree they need to do something about.”
As it happens, practitioners at the larger fund management firms that are already using the Convergence data for their own purposes have become an informal research and development group for the firm. They tend to ask Convergence to develop on their behalf the ideas that occur to them, and they invariably permit the firm to turn the results into a commercial service. “A lot of users want to create their own custom universes to benchmark themselves,” explains Phinney. “They might, for example, want to split credit funds into four buckets. We empower our users to create their own view of the world. If they want something more permanent, they ask us to create it, and then pay for it, and they do not object to us turning that into another product for sale.”
On the other hand, even pure database users are expected to respect the intellectual property and patent rights of Convergence in all of the services that the firm has developed. Convergence has trademarked 23 of what it calls “artefacts,” to signify that they incorporate the knowledge and experience of the principals as well as data extracted from public sources and analysed using well-known mathematical algorithms. For example, Convergence has developed algorithms that can distinguish between the needs of straightforward equity long/short funds with a single strategy, prime broker and administrator trading in cash instruments only from a single location, and a multi-strategy manager operating with multiple service providers from several locations, with complex fund structures, and making extensive use of derivative instruments.
“What we do is extract, transform and load the data,” explains Phinney. “We do not think the extraction process is proprietary. It is hard, but have figured it out, and other people can do that as well. Nor is loading the data difficult. `Transformation’ is what is hard. It is easy to discover that the name of a bank is spelt many different ways in Form ADVs, and we have normalised them. It is a simple syntax correction. It is once you start enriching the data, in the sense of creating new content and new artefacts, and combining them in different ways, that you create intellectual property. If that property is properly protected, no one else can create it and sell it or use it. We think we have created an effective intellectual property moat against competitors coming in and trying to do exactly what we are doing. We know there is at least one big administrator out there trying to do this. There are other groups out there already who are extracting the data, loading it and shipping it to customers, but that is all they do. You have to knock yourself out to make use of it. They are missing the enriched artefacts we offer.”
Distributing those “artefacts” is expensive and time-consuming, which is why Convergence is now seeking financing to scale the business. “We are selling strategically to the thought leaders in the hedge fund industry and their service providers, and proving the validity of our concept,” says Phinney. “We now have clients, revenue, pipeline and product. We want the financing to take us to the next level. We believe we have the potential to build a $30-50 million per annum revenue business over the next ten years, with very high margins, but you have got to have an organisation that delivers sales, distribution and service. Funnily enough, those are actually the easier parts. We have already spent the money on the harder parts, which are designing and building the technology and bringing it to bear on this data set, which it took us 18 months to really nail down. Now we have done that, the number of products we can create is unlimited. There is so much information in the data set, people are asking us different questions about it all the time, and we are able to find them the answers.”
Ultimately, the principal limit on the expansion of the Convergence data business is the amount of data that is publicly available, especially in jurisdictions outside the United States, where privacy protection laws tend to be more stringent. However the opportunity created by the data now being collected by regulators all over the world, most of it almost completely untouched by the resource-constrained regulatory bodies that collect it, is not lost on the principals at Convergence. They have already advocated public disclosure of the contents of Form PF, with a six to 12 month delay to counter objections from fund managers to the release of commercially or risk sensitive information. “The SEC is very fond of what we are doing, chiefly because they recognise that we are educating the market,” says Phinney. “Hopefully, regulators in other jurisdictions will come to appreciate the benefits of what we do as well. Regulators might one day be among our biggest clients.”
