This site uses cookies to provide you with a more responsive and personalised service. By using this site you agree to our use of cookies. Please read our privacy statement for more information on the cookies we use and how to delete or block them.

The Valuation Landscape in 2020

15 May 2020

Challenges and New Developments

by Simon Greaves & Martin Anastasov.

The financial landscape in the past few years has been changing dynamically and quite significantly in some areas. The new financial reality after the credit crisis of 2008 has been one of increased regulatory pressures and a re-definition of some traditional, long-held beliefs about the roles of market participants. Banks have been regulated away from high-risk activities, such as proprietary trading, and have focused on a new role centred on financial intermediation. Asset managers have attracted record levels of assets under management, mostly concentrated in passive investment vehicles. Equity markets have been buoyed to record highs by persistently low interest rates during a rare twelve-year bull market. These market developments have led to some unique valuation challenges as well.

Regulation – a strong driver for valuations

Regulation has been at the core of the increased demand for more rigorous valuation processes. New regulations, such as the fundamental review of the trading book (FRTB), have led to an increased emphasis on valuation processes and internal models, since they are the main building blocks, on which risk and capital adequacy measures depend.

Furthermore, regulation has also posed some new challenges in terms of valuation approaches. The transition from LIBOR to OIS rates, such as SOFR, SONIA and EONIA, has meant a major overhaul of internal pricing models at many financial institutions. To a more limited extent, the recent change in convention in the pricing of EUR and GBP-denominated swaptions from cash-settlement to physical delivery quotations has also increased the demands on valuation models, which would have to be extended to also model the basis between the cash and physical annuity measures used for pricing. Even though such a change in convention might look limited in terms of impact, it does have far-reaching consequences for valuations, as many more complex products, such as constant maturity swaps, have traditionally been priced and hedged by means of replication with swaptions.


The period between 2016 and 2018 was characterised by unusually low levels of realised volatility. This calm environment was disturbed by sharp increases in volatility towards the end of 2018 and in the current challenging environment. Such extreme changes in volatility regimes underlines the importance of modelling stochastic volatility in valuation approaches and capturing fat-tailed distributions adequately. It could potentially also lead to the more widespread usage of more innovative approaches[1]. Some institutions have taken the first steps towards the introduction of alternative risk management and graph modelling techniques[2] as a supplement to more traditional quantitative measures, such as VaR and expected shortfall.


Both the credit crisis of 2008 and the current challenging environment underlined the importance of liquidity and reminded market participants how quickly it can disappear. Including provisions for illiquidity is essential, yet few valuation models formally consider the issue of liquidity. To an extent this appears justified as illiquidity provisions are notoriously hard to make. Faced with scarce market data, valuation experts have traditionally relied more on judgment and experience. However, new developments in the fields of data science, machine learning and artificial intelligence may better inform such illiquidity provisions by utilising richer and less structured alternative data sources.

Increasing complexity and uncertainty of valuations

The modern financial landscape gives rise to valuation problems of increasing complexity and makes the valuations more uncertain and dependent on intangibles or unobservable data. Several technology start-up companies have managed to sustain billion dollar valuations despite failing to record accounting profits. Even large financial institutions recognise that additional layers of complexity can easily be introduced and skew traditional valuation models. Despite the widespread rise of collateralization, many market participants still do not fully account for the funding cost of uncollateralised derivatives (i.e. unfunded FVA) or the switch optionality inherent in their Credit Support Annex (“CSA”) agreements. Even something as simple as a bid-offer spread may introduce uncertainty in a valuation methodology and can question whether an asset can really be realised at the price suggested by a standard valuation model.

The increasing complexity of exotic financial products has also contributed to the uncertainties in valuation. Frequently, it is difficult to pinpoint the source of valuation disagreements for exotic products. Differences in valuations may be attributable to a flawed modelling methodology for the product, but they may also arise from incorrect model inputs, such as inappropriately marked volatility surfaces, or from the choice of simpler instruments used to replicate and hedge the exotic product.


There isn’t a single nor straightforward solution to many of these issues. However being aware of your own portfolios and the potential pitfalls that lie within is critical in order to be able to proactively address the areas that cause the greatest risk and the possibility for not just unrealised but real cash losses

[1] Examples include multifractal models or modelling volatility by means of truncated Levy flights.

[2] Such as Bayesian risk management.