ETFs, automated trading and financial market efficiency

This section will begin by outlining the mechanisms through which information is incorporated into financial market prices. It will then proceed to analyse the potential effects upon these mechanisms brought about by the widespread utilisation of passive investment vehicles such as ETFs, and by the introduction of certain FinTech tools, in particular automated trading. Given the spectacular growth of ETFs and the burgeoning FinTech sector, the systemic risks associated with these markets have been comparatively under-studied. Many regulatory analyses have pointed to the role of ETFs and algorithmic trading in accentuating steep and rapid but infrequent and temporary drops in exchange prices -so-called ‘flash-crashes’ - but systemic risk is frequently regarded as tangential to these episodes. Because of this, few enquiries have concentrated on the potential for ETFs to exacerbate, or cause, broader and enduring systemic risk events. Similarly most recent studies on the risks from FinTech are confined to examining risks at the micro-level; for example, the risk of losses incurred by individual consumers from faulty automated advice, privacy concerns or cybersecurity risks.[1]

The neglect of systemic risks from these technologies, on the face of it, is not surprising. By increasing the speed of transactions, removing intermediation costs and increasing competition, financial innovations such as ETFs and algorithmic trading are assumed in almost all circumstances to augment market efficiency. Because the most popular ETFs are designed to mimic a wide market portfolio - often an entire securities or bond exchange -there is normally no question of relative price efficiency that is, the question of whether collectively, investments exhibit efficiency in relation to one another. Hence, studies on ETFs frequently find that inclusion of a security in an ETF improves its informational efficiency.

Despite these findings however, the effects of higher investment volumes made through passive vehicles may impact the informational efficiency of securities prices in the aggregate, through two distinct, but linked, phenomena. First, as capital allocators, ETFs will amplify existing market movements. ETFs are at root pure momentum traders and will buy or sell securities depending upon prevailing market conditions, exacerbating any mispricing which occurs. Because ETF trading requires commensurately lower capital commitments to execute, and the underlying securities of an ETF are less liquid, taking large directional bets on an index is facilitated by ETF vehicles.

Second, as ETFs and certain forms of automated trading continue to proliferate, informed trading - as a mechanism for correcting mispricing through arbitrage and other techniques - will become less likely to deliver returns because the mass of capital flows that such arbitrage strategies must correct are potentially much larger. Such processes are already recognised in algorithmically traded markets; informed traders have fewer incentives to participate in algorithmic markets and to correct informational deficits. As passive investment techniques continue to dominate active trading, price discovery processes will be undermined, making asset value swings more likely.

Information costs and market efficiency

Classic expositions of market operation centre on information problems as a source of inefficiency. Information asymmetries and a lack of information dispersal may result in imperfect market functioning and these inefficiencies have been central to recent stock market collapses and financial crises.

Readers will be familiar with Fama’s efficiency form trichotomy: ‘strong’; ‘semi-strong’ and ‘weak’. Each of these forms assumes that markets are fundamentally efficient, but each to different degrees, based upon the speed through which information is incorporated into

Exchange-traded funds (ETFs) and FinTech 235 prices. The strong-form EMH hold that securities prices reflect all available information, both public and private, and therefore no trader may earn excess returns even where they trade on private information. The semi-strong version holds that securities prices reflect all publicly known information and adjust instantaneously to the production of new information. Profit from trading on private information is therefore possible, but the window for profit is tightly constrained. The weak version of the EMH holds that prices reflect only historically relevant information to the securities concerned. Therefore, traders may make profit on private information-gathering, but cannot outperform the market for consistently long periods.[2] The importance of the EMH to the study of financial markets is ‘its prediction that, even though all information is not immediately and costlessly available to all participants, the market will act as if it were’.

  • [1] Indicatively, see M Demertzis, S Merler and G B Wolff, ‘Capital Markets Union and the Fintech Opportunity’ (2018) 4 J Fin Reg 157. 2 L Glosten, S Nallareddy and Y Zou, ‘ETF Activity and Informational Efficiency of Underlying Securities’ (forthcoming 2020) Man Sci https://pubsonline.informs.org/doi/10.1287/mnsc.2019.3427. 3 D Israeli, C Lee and S A Sridharan, ‘Is There a Dark Side to Exchange Traded Funds? An Information Perspective’ (2017) 22 Rev Acc Stud 1048. 4 G Akerlof, ‘The Market for Lemons: Quality Uncertainty and the Market Mechanism’ (1970) 84 Q J Econ 488. 5 B Holmstrom, ‘Understanding the Role of Debt in the Financial System’ BIS Working Paper No. 479 (January 2015) https://www.bis.org/publ/work479.pdf. 6 E F Fama, ‘Efficient Capital Markets: A Review of Theory and Empirical Work’ (1970) 25 J Fin 383.
  • [2] Fama (n 39). 2 Gilson and Kraakman (n 16) 552 [emphasis in original]. 3 Gilson and Kraakman (n 16).
 
Source
< Prev   CONTENTS   Source   Next >