2nd CEPR-Imperial-Plato Market Innovator (MI3) Conference 2018
The Sound of Many Funds Rebalancing
Alex Chinco (University of Illinois at Urbana-Champaign)
Vyacheslav Fos (Boston College)
Alex Chinco is an Assistant Professor of Finance at the University of Illinois at Urbana-Champaign. His main research areas focus on asset pricing, behavioural finance and real-estate finance. He holds a PhD in Finance from New York University, and a BA in Economics from the University of Chicago.
Vyacheslav Fos is an associate professor at the Carroll School of Management, Boston College. His research areas are related to financial markets, corporate governance and corporate finance, focussing on information transmission in financial markets, stock liquidty, activist shareholders and consequences of share buybacks.
About the Paper
Alex and Vyacheslav’s paper proposes that long rebalancing cascades generate noise in financial markets. Using a random-networks model it’s possible to show that, when funds follow many different threshold-based trading strategies, a change in stock A’s fundamentals can trigger long rebalancing cascades that affects demands for unrelated stocks.
In a large market it’s computationally infeasible to predict whether a long rebalancing cascade will result in buy or sell orders for unrelated stocks, though it is possible to compute stock Z’s susceptibility to erratic non-fundamental demand shocks.
Without noise, traders will never be able to take advantage of under-priced stocks, as other traders would replicate your trades, and nobody would be willing to sell you trades at old prices. Noise makes markets as different people buy and sell different stocks for apparently erratic reasons. It allows traders to re-balance their books without notifying other traders of their new positions. In Fischer Black’s words “noise makes financial markets possible.”
So where does this noise come from? Standard answers is that it comes from individual investors and their demands. However, the diminishing importance has steadily declined, while the noise remains. Motivated by this decline in traders – but no corresponding drop in noise – Alex and Vyacheslav propose an alternative noise-generating mechanism: computational complexity and index-fund rebalancing cascades.
They begin their analysis by presenting a model where, because there are so amny index funds, a small change in stock A can affect Stock Z, by way of stock B, C, D… etc. While it’s possible to determine if a stock will be affected, the problem of determining how the stock will be affected is computationally impossible.
This paper from Alex and Vyacheslav proposes an explanation for seemingly random demand shocks in financial markets. They are able to empirically prove that index-fund rebalancing cascades actually generate noise in real-world financial markets using data on the end-of-day holdings of ETFs.
By showing why it’s computationally intractable to predict ETF balancing cascades, it is possible for researchers to identify other situation where the same logic holds. Their theoretical model also applies to any other group of funds following a wide variety of threshold-based rebalancing funds.
In addition, they are able to estimate the amount of additional liquidity that is associated with stocks having more neighbours.