Skip to main content
Skip table of contents

Q4 2019

This release updates EDHECinfra’s equity and debt indices as of Q4 2019.

The Q4 2019 release of the EDHECinfra indices includes a number of improvements and bug fixes that have improved the stability and accuracy of results. They include better and more input price data from secondary equity markets including in the previous years, where more price data is now available and has impacted the calibration of our risk premia model for unlisted infrastructure equity. We also integrated more interest rate data, including for countries from it was not previously available. As a result, the improved risk premia estimates and the more accurate use of long-term bond yields for risk-free rate curves have changed the results presented in Q3 2019 to a point. This is normal as the development of the index research continued in 2019. From 2020, We will be ‘freezing’ the historical index results to ensure stability and consistency over time. The index freeze will have an 18-month (six-quarter) lag due to the fact that some of the relevant financial and price data can take up to 18 months to be revealed and integrated in the estimation of the risk premia and cash flow forecasts. During that 18-month window, all results can be considered as ‘best estimates’ of the risk and returns of the relevant index constituents.

  1. Model updates: 

    1. Based on consistent feedback, the debt index constituents are now incorporated at an instrument level. Each senior debt instrument is priced individually and is now a constituent in the index. This change brings EDHECinfra’s methodology into line with other debt indices in the industry. The new approach differs from previous versions when all senior debt of a company was considered as one unit of account.

  2. Calculation updates: 

    1. Some of the senior debt instruments draw down over time and, hence, EDHECinfra has improved the price return calculation to reflect the outstanding principal at every point in time.

  3. Data changes: 

    1. More observed equity transaction prices have been added to improve the accuracy of calibrating risk premiums to market prices. An extensive validation of historical transactions has also refined the quality of past data.

    2. More than 200 company accounts have been updated in this quarter.

    3. EDHECinfra has increased the coverage of observed risk-free rates in terms of maturities and countries, using government bonds wherever possible or other proxies. This has enhanced the forecast of interest rate curves at different points in time in the history, which is one of the key elements of EDHECinfra’s asset pricing approach.

  4. Point in time revenue forecasts: 

    1. Our index methodology captures the entire history of all the included companies in the universe. However, in order to price them correctly in the history, we have had to generate multiple revenue forecasts at different points in time to capture the evolution of their expected dividends and any shifts in the industry/market conditions that might impact expected cash flows (e.g. trend in energy prices, change in regulation etc). For example: the Spanish toll road sector went through a significant downturn from 2012-16 period. This update of our data ‘at each point in time’ has been an incremental process and is now completed.

  5. Term structure of interest rates: 

    1. We have significantly increased our coverage of the government bonds across maturities and countries. Due to this, we are now able to estimate the entire yield curves at every point in time with a much better precision. Amongst other things we now discriminate between European countries when we used to have a single ‘Euro’ interest rate. And in the countries where the government bonds do not trade often, we now rely on some other proxies such as interbank rates. This is one of the most significant changes in our index for three reasons:

      1. We rely on the DCF method for valuation and use the term structures for discounting the cash flows

      2. Interest rates also act as the base rate in determining interest payments of some floating rate debt instruments

      3. Interest rates are also used as factors in our asset pricing model 

  6. Calibration of the asset pricing model:  

    1. We rely on secondary market transactions to calibrate the asset pricing model. This allows us to estimate valuations in line with the broader market which is also a requirement for mark-to-market treatment per IFRS13 guidelines. In 2019, we have devoted significant resources in cross-checking and correcting the observed secondary prices that we had captured from different sources. For example: in some cases enterprise value was given as equity value. This has led to changes in model calibration. This cleanup of reported market prices has significantly improved the precision and robustness of the risk premia model.

  7. New information:  

    1. We always try to capture the best available information for new and existing companies in the universe. This leads to changes in the history of the index since the new company is added along with its entire history or an existing company might have reported updated accounts. To counter this, we will start to freeze the index history from Q1 2020 and allow only changes in the last six quarters to reflect the newly reported financials of the underlying companies.

  8. Instrument level pricing (only applies to debt index):  

    1. We used to recognise all the senior debt on a company’s books as one unit of account. The pricing model was also based on this bundled debt and it also represented the index constituents. However, during the beta phase last year, we received feedback from numerous clients to change this to individual instruments since that’s what they essentially invest in. This has been implemented in Q4 2019 and it implies that the debt index is entirely different now. This approach has also allowed us to build custom indices tailored to the specific properties of the debt instruments such as RPI-linkage, currency of the issue etc.

JavaScript errors detected

Please note, these errors can depend on your browser setup.

If this problem persists, please contact our support.