{"id":38362,"date":"2014-11-21T01:38:40","date_gmt":"2014-11-21T00:38:40","guid":{"rendered":"http:\/\/beta.next-finance.net\/strategie\/a-smart-selection-process-for-etfs\/"},"modified":"2019-12-30T23:12:44","modified_gmt":"2019-12-30T22:12:44","slug":"a-smart-selection-process-for-etfs","status":"publish","type":"post","link":"http:\/\/beta.next-finance.net\/en\/strategie\/a-smart-selection-process-for-etfs\/","title":{"rendered":"A smart selection process for ETFs"},"content":{"rendered":"

Until recently, there was no satisfactory scientific answer to this question, but Lyxor\u2019s research teams have changed all that.<\/strong> <\/em><\/p>\n

Traditional selection tools are not suited to ETFs<\/h2>\n

Institutional investors, such as pension funds, insurance
\ncompanies and balanced fund managers who rely heavily
\non trackers, have to bear this in mind when choosing the
\nbest ETFs for their portfolio.<\/p>\n

They face a major challenge. The analysis tools traditionally
\nemployed for active fund selection are not suited to selecting
\nETFs. In its simplest expression, active fund selection
\nis based on one fundamental criterion: the information
\nratio. This indicator measures a fund\u2019s return relative to its
\nbenchmark index, taking into account the relative risk taken
\ncompared with said index. However, using the information
\nratio to compare ETFs has a limitation which totally invalidates
\nthe analysis.<\/p>\n

As a reminder, the information ratio is the ratio of
\noutperformance (alpha) to tracking error (TE) volatility. In the
\ncase of ETFs, which replicate the index, the outperformance
\nand TE figures are very low and the information ratios are
\ntherefore extremely sensitive, making traditional analysis
\nirrelevant. Moreover, a product with a very low tracking error
\ncould easily be mistakenly ruled out on account of a weak
\ninformation ratio, despite excellent index replication. Lastly,
\na product underperforming very slightly and with a very low
\nTE will be ruled out on account of its negative information
\nratio, whereas a product with marginal outperformance will
\nbe retained even with a high TE.<\/p>\n

<\/p>\n

A new framework for measuring ETF efficiency…<\/h2>\n

Basically, from an investor\u2019s point of view, a good ETF
\nshould maximise the chances of replicating index returns.
\nThe fund must also display a low bid-ask spread in order to
\npreserve the profit on the trade. In fact, a suitable analysis
\nframework for ETF selection derives from three fundamental
\nparameters: an estimate of the performance gap between
\nthe fund and its benchmark (i.e. the tracking difference);
\nthe volatility of this performance gap (i.e. the tracking error);
\nand the difference between the buy and sell prices (i.e. the
\nbid-ask spread or \u201cliquidity spread\u201d). By applying Value at
\nRisk (VaR) \u2013 now the most widely used measurement of
\nrisk \u2013 to these three parameters, it is possible to accurately
\nmeasure the efficiency of an ETF.<\/p>\n

In concrete terms, by using a one-year Gaussian VaR at
\na 95% confidence level, this efficiency measurement is
\nexpressed as follows:
\n
Efficiency = Tracking Difference \u2013 Liquidity Spread \u2013 1.65 x Tracking Error <\/strong><\/p>\n

For instance, if the efficiency of the ETF is equal to -50 bps, the
\nprobability that the investor faces a relative loss with
\nrespect to the index larger than 50 bps is exactly equal to 5%.<\/quote><\/p>\n

Using this risk measurement therefore makes it easy to
\ncompare the efficiency of two ETFs. This model, which is
\ndescribed in detail in a research article available online at
\nhttp:\/\/ssrn.com\/abstract=2212596<\/a>, and which was
\nalso published in Journal of Index Investing, works
\nas follows. The higher the outperformance of an ETF, the better
\nits efficiency, while a wider bid-ask spread makes it less
\nefficient. Equally, higher tracking-error volatility increases
\nuncertainty and thus makes the ETF less efficient. It is
\ninteresting to note that comparing ETFs by efficiency rather
\nthan by individual criteria available to investors (such as
\noutperformance, daily spreads or volatility) produces
\ndifferent results. Therein lies the strength of this synthetic
\nindicator. <\/p>\n

ILLUSTRATION 1 – MSCI Emerging Markets Index Efficiency Indicator<\/strong> <\/em>
\n\"illustration_1_-_msci_emerging_markets_index_efficiency_indicator.jpg\"<\/p>\n

…taking account of institutional concerns<\/h2>\n

In the basic calculation for the efficiency indicator, we are
\nusing the best limit order spread for the liquidity spread.
\nYet, efficiency measurements can be refined in order to
\nbetter reflect the daily realities of institutional investors. Such
\ninvestors\u2019 orders can involve amounts running into tens of
\nmillions of euro. Even when split, they generally cannot be
\nexecuted at the best limit. Based on order-book historical
\ndata, it nevertheless remains possible to measure the
\naverage bid-ask spread at which a given notional amount will
\nbe executed. This \u201cliquidity spread\u201d can then be reintroduced
\ninto the calculation of the indicator. Interestingly, depending
\non whether the analysis uses a notional amount of 100,000,
\n1 million or 2 million euro, the efficiency measurement
\npinpoints different ETF providers for the same index. This
\nhighlights the importance of ETF liquidity for investors.<\/p>\n

ETF providers eager to bring the highest level of service to
\ntheir investors must endeavour to improve the liquidity of their
\nproducts in order to minimise the bid-ask spread. Having a
\nlarge number of market makers is important for ETF liquidity.
\nAt Lyxor, everything is done to ensure that each fund is
\nfollowed on average by nine market makers. In addition, in
\norder to allow each investor to use the model presented
\nin this article and select ETFs efficiently, it is essential that
\nETF providers publish material on the tracking-error volatility
\nof their funds, in line with recommendations provided by
\nthe European Securities Market Association (ESMA), and
\nthat stockbrokers develop suitable statistical measures for
\nunderstanding ETF liquidity spread.<\/p>\n

ILLUSTRATION 2 – Euro STOXX 50\u00a0 Index Liquidity Spread<\/strong> <\/em>
\n\"illustration_2_-_euro_stoxx_50_index_liquidity_spread.jpg\"