Tackling the Look-through Challenges with In‑Memory Computing

Look Through ApproachIn the April 2015 edition of its Global Financial Stability Report, the IMF raised concerns about potential financial stability risks posed by the asset management industry, calling for regulatory scrutiny on a sector which intermediates 40% of the world’s financial assets. Whether under regulatory or client pressure, asset managers should consider the technology implications of a greater transparency in risk reporting sooner rather than later. This post will delve into the implications of the look-through approach from a data management standpoint, building the case for the use of modern in-memory aggregation technology to process massive amounts of highly granular data.

A definition of look-through

Look Through ApproachThe requirement for a more granular view of risk is one of the founding principles of the look-though approach which insurers are expected to adopt as part of Solvency II compliance. In its paper A Look-through Approach to Managing Investment Assets, Accenture explains the look-through principle as follows; “The look-through principle enables a refined level of asset monitoring and risk management, based on the availability of more detailed information on the exposure of investment assets and pooled funds to market risks.”

Why should asset managers care?

Asset managers consider that investing a fund in other funds is a good tactic, offering a global reach to specific strategies or specific asset classes. It demonstrates their ability to access a very wide pool of funds, for the benefit of their clients. The issue, however, is that the risk created by investing a fund in other funds cannot be captured by looking only at the global risk figure of a portfolio. A global line will not provide any visibility into the composition of each fund, let alone into the correlation between each component. Unwanted concentrations within a given fund or between funds will not be obvious at first sight if you look only at the global risk figure of the portfolio. For example, if you would like to calculate the value at risk (VAR) of the portfolio, then you need to decompose each fund into its most granular constituents.

In fact, the look-through approach is tackling the requirement for a greater level of transparency in the reporting. By adopting a look-through approach, asset managers are able to answer questions such as what is the contribution of a specific asset to the return of my investment portfolio? How is this asset contributing to risk of the portfolio? What is the Marginal VAR of a given specific asset? And so on.

The data volume headache

The data volume headacheThe level of granularity encouraged by the look-through principle translates into an explosion in data volumes, as each and every individual position should be taken into account when assessing risk. More data can turn the reporting of pooled funds into a daunting task for insurers and asset managers alike, and put great strain on existing data management infrastructures. Any firm still using legacy data warehouse technology will struggle with risk analytics, experiencing slow response times because of the sheer volume of individual positions needed to deliver the required level of granularity. Come to think of it: gathering and processing data about each individual position from up to hundreds of asset managers will be a daunting exercise for an insurance firm. Without the proper technology, the cost of internalizing teams of data analysts dedicated to that task would simply be too high. It is an equally painful undertaking for large asset management firms who manage “funds of funds”, using complex investment strategies with multiple levels of funds composition.

With clients increasingly requesting their asset managers to produce reports on a much more frequent basis and with a higher level of granularity, in-memory analytical platforms such as ActivePivot are a great response to the data management challenges that relate to the look-through principle. Because ActivePivot is natively designed to store and access huge volumes of data in-memory whilst combining transactional and analytical processing, it enables instant what-if analysis. As soon as you change parameters, complex metrics are recomputed on the fly, allowing you to simulate the impact of multiple fund rebalancing scenarios. As a result, risk reporting and analytics are becoming a highly interactive undertaking that supports operational decision making.

Looking beyond the regulatory mandate

Although the look-through approach is driven by a regulatory mandate, it would be a mistake not to turn it into an opportunity. To quote the Financial Times in a recent article “Assessing Solvency II’s ‘look-through’ approach”, dated February 15th: “Getting ready to execute on the look-though approach represents an opportunity for asset managers. Solvency II presents an opportunity for insurers and their asset managers to improve transparency and decision making, to optimise exposure to rewarded risks, enhance investment performance and achieve a higher return on capital”. There is a clear case for asset managers to differentiate themselves from competitors by offering look-through as a value-add service to their clients. By adopting a purpose-fit data management infrastructure, large asset management firms would not only be able to deliver on the expected level of granularity, but also propose innovative analytics services around risk management that would in turn, increase customer loyalty and translate into incremental revenues.

The look-through approach is a good example of how in-memory technology can be used to turn a regulatory mandate into a business opportunity.


- IMF: http://www.imf.org/External/Pubs/FT/GFSR/2015/01/pdf/c3.pdf
- Financial Times: http://www.ft.com/intl/cms/s/0/833ed52e-b6bb-11e4-a5f2-00144feab7de.html#axzz3Xx2iE0p4

Like this blog?
Follow us on LinkedIn
to stay up-to-date with the latest blog posts!

This entry was posted in Big Data Analytics, Finance. Bookmark the permalink.

Comments are closed.