Financial intermediaries and contagion in market efficiency:
The case of ETFs

  • Claire Yurong Hong, Frank Weikai Li, Avanidhar Subrahmanyam
  • SSRN Working Paper
  • A version of this paper can be found here
  • Want to read our summaries of academic finance papers? Check out our Academic Research Insight category

What are the Research Questions?

In this research, the authors explore the role of financial intermediaries in contagion or comovements in pricing efficiency.  Specifically, lead market makers (LMMs) like Goldman Sachs, Cantor Fitzgerald, RBC Capital Markets, and others, have funding constraints that may influence their ability to accurately price ETFs and cause contagion in the pricing of financial assets.  The market for ETFs provides an attractive laboratory to test this hypothesis because of the wide diversity and high use of the product. The following questions were analyzed using the universe of ETFs listed on US exchanges over the period from 2012 to 2020.

  1. Why use ETFs to study this proposition?
  2. Do capital constraints influence pricing efficiency across financial assets?
  3. What about spillover effects?

 What are the Academic Insights?

  1. The ETF market is appropriate for this topic for a number of reasons. The market for ETFs allows the identification of capital constraints that are specific to the ETF as separate from aggregate funding constraints. The authors define the capital constraint specific to LMMs as that contributed by the creation and redemption pressures for maintaining the ETF. This is an important feature when financial intermediaries are marginal investors, as any constraints that are specific to the intermediary will exert a larger impact on prices.  Further, unlike other financial assets, the price of ETFs is clearly defined as the exact replication of the prices of the underlying assets.  Finally, the size of the ETF market makes it economically significant, therefore questions of efficiency or inefficiency are of interest.

  2. YES. Using the market conditions surrounding the COVID-19 sell-off to test the main hypothesis, confirming results are documented in Table 8 below. During the period, from January to June 2020, unparalleled levels of pricing gaps occurred where ETFs were mainly priced at a discount. This was true, especially for fixed income(FI), which is the class of ETFs tested here. Note the coefficient on the interaction between FI weight and Covid is 9.35 (t=2.88).  How should one interpret this outcome?  The authors explain the economic significance: “a non-FI ETF managed by an LMM with a 75% weight in FI ETFs exhibited a premium 4.68 bps greater than ETFs managed by an LMM with only a 25% weight in FI ETFs.” Apparently, the negative impact of Covid on the FI ETF market leads to larger ETF gaps, 4.68 bps in this case.

  3. YES, if they share common financial intermediaries. Here’s how it works:  LMMs are obligated to maintain the “Law of One Price”, that is, the replication of prices of ETFs with the net asset value (NAV)  of the underlying assets. When prices diverge in the direction of either a premium or discount, between the ETF price and its NAV, LMMs take positions designed to arbitrage the divergence.  A long position on the relatively underpriced asset is paired with a short position on the relatively overpriced asset.  This activity absorbs capital as the LMM directs more funds to the ETF in order to close the pricing gap.  The unintended consequence is to make less capital available to other ETFs managed by the same LMM and promote contagion and spillovers.

Why Does it Matter?

The question of price contagion across LMMs and the assets they mediate is important as it relates to issues of market efficiency. Of more importance is the impact the contagion may have on investors using ETFs to build portfolios, in order to obtain diversification without much difficulty.  However, the contagion documented here may have adverse effects on the benefits of that very same diversification.  It should be noted however that the largest impact was on ETFs which held illiquid assets. Imagine the result, if ETF returns are strongly correlated with the returns of other asset classes and this occurs during periods when intermediaries have serious constraints on funding. Diversification could be drastically curtailed when the comovements of pricing are strongly correlated.

The Most Important Chart from the Paper

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained.  Indexes are unmanaged and do not reflect management or trading fees, and one cannot invest directly in an index.


Capital constraints of financial intermediaries can affect liquidity provision. We investigate whether these constraints spillover and consequently cause contagion in the degree of market efficiency across assets managed by a common intermediary. Specifically, we provide evidence of strong comovement in pricing gaps between ETFs and their constituents for ETFs served by the same lead market maker (LMM). The effects are stronger for ETFs that are more illiquid and volatile, when the underlying constituents of the ETFs are more costly to arbitrage, and for LMMs with more constrained capital. Using extreme disruptions in debt markets during COVID-19 as an experiment, we show that non-fixed income ETFs serviced by LMMs managing a larger fraction of fixed income ETFs experience greater pricing gaps. Overall, our results indicate that intermediaries’ constraints indeed influence comovements in pricing efficiencies.