Beneath Calm Waters:
The Challenges of Model Portfolio Management
In its simplest form Model portfolio management involves constructing and managing a portfolio representing a specific investment strategy, theme, or risk profile. These portfolios are designed to simplify the investment process for advisors and provide investors with a diversified and professionally managed investment solution. They are, in effect, a template that the RIA (Advisor), or the platform, can then populate with investments. In this way, a platform can offer portfolio models from a range of investment managers in one place to their clients, providing flexible solutions, choice of investment manager and cost efficiency all in one.
Whilst the use of model portfolios has continued its strong growth globally this rapid growth has brought fundamental issues which need to be resolved, namely model proliferation, growing fragmentation, and lack of standardization. This, coupled with a high degree of manual processes, represent clear challenges for the industry's continued growth. In the following sections, we aim to address these challenges using practitioner insight coupled with case studies to add depth and relevance, before looking at possible solutions and best practice suggestions.
Model Portfolio Management: Under the surface
Implications of Model Proliferation
One of the critical challenges is the proliferation of platforms and portfolios. With numerous platforms and various portfolios available, it becomes increasingly difficult for investors to navigate the options and select the most suitable ones for their needs.
But it's broader than simply investor choice; asset managers find that their suite of models is growing from the 10s to 100s and beyond as each platform demands a slightly different model with a different investment universe. This brings us to fragmentation.
Fragmentation in the industry and its challenges
Fragmentation means one Advisor or platform has a different set of underlying assets available. Therefore, they need a different model, or the asset manager must translate a core model to a bespoke one based on the end requirements. This can be as mundane as different limits on cash or as significant as lacking an investment vehicle.
Proliferation has bred fragmentation in the industry, making communication and coordination between different platforms and asset managers' portfolios a significant challenge. This threatens one of the central benefits of Model Portfolios, namely, scalability. The lack of standardized interfaces and processes further exacerbates this issue, as much work still needs to be done manually on spreadsheets. In addition to inefficiencies, it also increases the risk of errors and inaccuracies in portfolio management.
The importance of standardization in the face of manual processes and risks
There needs to be more standardization across the industry. Currently, different end users require different data formats and operate within diverse and often non-overlapping investment universes. One of the main benefits of model portfolios should be efficiency and scalability. This is challenging without standardization.
For example, one asset manager may construct and run several models for clients' use cases. These are then pushed out to Advisors or platforms, with each platform or Advisor requiring a slightly different flavour of the model. For example, bespoke identifiers, investment universes or construction rules. The lack of standardization in model construction and communication, not to mention the execution, is an enormous drawback to continued success. A direct result of this is the need to resort to manual processing, with spreadsheets often sitting at the heart of managers' portfolio construction processes.
Standardized interfaces and processes could, in a best-case scenario, enable seamless communication and coordination between platforms and portfolios, enhancing efficiency and reducing the risk of errors. Additionally, standardization would facilitate comparing and evaluating different portfolios, providing investors with a clearer picture of their options. While one never achieves the best-case outcome, a move even part of the way in this direction would be beneficial. A first step here is phasing out manual spreadsheet-based processes.
The dominance of manual processes in model portfolio management poses risks to everyone involved. While asset managers can and should move towards more automated portfolio construction and risk frameworks to comprehensively tackle the issue, we need platforms to get on board; for example, accurately forecasting income and determining when a trade is complete can sometimes be a challenge requiring manual workarounds. Also, the distribution of models to advisors should be based on something more reliable than spreadsheets. This all involves investment and pause for thought, but in the UK, there may be an additional push factor in legislation.
UK Consumer Duty legislation
In July 2023, Consumer duty legislation was introduced in the UK, making some form of Value Assessment a regulatory requirement for Model Portfolios. Providers must now demonstrate how they will deliver good outcomes and value to clients. This has led companies to review and optimize product design, client engagement, and documentation practices.
While the ultimate impact is unclear, with asset managers still working out exactly what the regulation means for them, it would appear positive for advisors. Firms must now demonstrate the details behind their solution regarding service and value. Businesses must reassess fees to comply with fair value requirements.
Although Consumer Duty may disrupt Model Portfolios, it should make it even more appealing to advisors, allowing them to scrutinize expected performance versus cost. This is critical for investors in our current high inflation, high-interest rate, and de facto high cost of living environment. One solution to the challenge of consumer duty legislation would be to build ex-post and ex-ante risk capabilities along with scenario generation to the Model Portfolios portfolio construction process. In the next section, we will explore these challenges and solutions further adopting the viewpoint of a UK practitioner; we will also use a case study based on a US asset manager to illustrate how some are already taking steps in this direction.
Practitioners Viewpoint: Beating against the current
Challenges
Ideally, a model portfolio solution should be automated on the asset manager side. A series of portfolios could focus on regions like emerging markets, China, etc., be thematic, focused on climate change, ESG, income, etc, or risk-focused, with conservative, balanced, or aggressive models. The beauty of the approach is that by building a process that begins with investment views, a small team can construct a series of flexible models based on one investment process and a set of ideas that serve different client needs.
A change in view should propagate through all models coherently. However, this is rarely the case; there are so many platforms or RIA's, and they all want something slightly different. Asset managers must either produce generic one-size-fits-all models or move in the other direction and increasingly customize for various end users. The former would allow more standardization and a seamless process, but it's unlikely to sell all that well; after all, one needs to stand out even in Model Portfolios. So, in reality, it's the latter approach that wins out. What this means in practice is a proliferation of models on the asset managers' side as each platform gets its own set of models that conform to its investable universe or risk/portfolio construction constraints. Managing these 100s or 1000s of model portfolios becomes its own cottage industry.
Asset managers do not directly invest capital in model portfolios, so they need to communicate that portfolio to the RIA or the platform; this is not a straightforward task as each platform has a slightly different architecture and way of consuming data, given the proliferation of models one cannot simply send a spreadsheet. The issue does not end with getting the model to the platform; you have ongoing changes and rebalancing to keep in mind.
As an example, let's say an investment firm has 1000 model portfolios, and its view on US bonds moves from neutral to underweight. That's a relatively simple change to the core model portfolio, and those changes can propagate quickly enough to variations on that core. If processes have been built robustly, it may even be able to automate the rollout to the 1000 models, but how are changes communicated to all of that investment firms' different platforms and advisors? Also, how is it possible to ensure the shift is executed correctly? And that the portfolios are correctly rebalanced once the bond weighting changes. This is a non-trivial issue requiring producers and distributors to work together. However, the benefits should be clear.
A related issue is the handling and attribution of income; in the UK, retirees now have more choices when accessing their pension, and many opt for a mix of traditional annuities and income- plus growth portfolios. A discussion of the retirement solution space is out of scope for this paper, but we should be able to observe how model portfolios could be helpful here. However, there is a problem: income handling and attribution by UK platforms is challenging. While the prediction, attribution, and distribution of income is a complex topic, it is clear that there is a receptive market for income-focused Model Portfolios.
These are all quite fundamental considerations. They arise in part because of the decentralised way the industry has evolved. This has positives as it drives innovation and competition, increasing creativity choice and lowering fees, but perhaps the time has come to innovate towards standardization and centralization.
Solutions
Standardization and centralization are vital to improving the current model portfolio landscape; fragmentation and proliferation are challenges that will persist in the medium term; therefore, solutions should initially focus on how processes can be built at the asset manager level. A 3-step approach could be applied: centralize, standardize and unify.
Centralize
A handful of core portfolios should be at the heart of a manager's universe of models. This is the model portfolio book of record (to use an accounting term). Changes to these core portfolios should propagate as appropriate through all the child portfolios. The entire ecosystem sits on one internal platform to ensure data consistency. Scalability and improved governance across the whole model suite
Standardize
Unify
Once an asset manager has built a standardized centralized model portfolio process, the next step is to communicate these models to various end-users; all the good work would be in vain if, at this point, one starts manually customizing models to fit different systems. Technology can sit in the middle and unify communication between the asset manager and multiple interested parties that would like to use those models but that all have differing investment universes and constraints.
The success of Model Portfolios now threatens to impact the approach's utility. The solution, in our opinion, is to embrace standardization; one way to do this would be to build unified interfaces based on third-party technologies; in simple terms, on one side, you have the asset managers, and on the other, the RIA's and platforms in the middle sits an interface which speaks both languages and can achieve the communication that is currently lacking. The following case study shows how one US-based client has used technology to move part way towards this.
Case study: Old dogs and new tricks
A US-based asset manager with a long history of mutual fund investing started to experience declining market share and falling AUM as RIA's moved away from investing in single mutual fund solutions towards constructing bespoke client portfolios. The manager had a growing Multi-Asset Solutions business (MAS) whose asset allocation process was partly driven by a series of model portfolios tilted around a central benchmark. At the same time, Model Portfolios solutions were taking off in the US and Asia with increasing adoption by RIA's; this provided an opportunity for the manager to leverage their existing strategic asset allocation (SAA) process to drive not one or two but multiple model portfolio's tailored to clients specific needs, enabling it to target Assets Under Advice AUA as well as the traditional AUM targets and generate new revenue streams.
Being a relatively late entrant to the market had its advantages, as the approach built could take some of the challenges we have mentioned into account. The process was to create a unified platform where changes in one or multiple core models would propagate downstream to child portfolios. This involved the integration of a tech solution from Jacobi.
The standardization and automation of this approach allowed the manager to quickly scale the process without any loss of fidelity or drop in governance. Another advantage of tech integration. was the ability to provide ex-post and ex-ante risk visualization and scenario analysis; this proved a valuable tool for both the portfolio construction process and the sales and distribution effort.
Risk and scenario analysis proved to be powerful sales tools, helping clients understand the utility of the proposed model portfolio and providing reassurance that the process behind them was robust and scalable. However, the usefulness extended beyond closing the deal; client updates benefited from clear performance attribution, and the ability to discuss the impact of future scenarios helped cement relationships and improve retention.
Conclusion and recommendations: We are going to need a bigger boat
The world of model portfolio management is something of a victim of its success; the rapid growth of the industry and increasing demand from end users and clients for more customization and diversification, alongside increasing regulatory oversight, means that the challenges we identified are unlikely to be solved organically in the near future, at least not on the platform or RIA side.
While it's easy for an investment manager to criticize the platforms and point out the deficiencies, there is plenty of work to be done to harmonize, centralize and standardize processes. We suggest the introduction of unified interfaces to facilitate communication with intermediaries without losing fidelity. But this does require some element of collaboration between producers and distributors.
Hopefully, in the paper, we have illustrated what some of the benefits of automation and collaboration might be. Still, it's worth keeping in mind the alternative: failure to tackle these issues could mean falling behind the competition as the Model Portfolios market evolves into a more unified standardized ecosystem that offers even more benefits to clients globally.
Model Portfolio Management Insights
.png?width=2000&height=1350&name=surge%20article%20(1).png "surge article (1)")