This paper shows that authorized participants’ (APs) regulatory constraints weakened the arbitrage relationship between bond ETFs’ primary market activity and their premia during the market turmoil triggered by the dash-for-cash episode in March 2020. Arbitrage activity weakened more severely when those APs with a history of creating or redeeming an ETF’s shares had low regulatory capital ratios, consistent with persistence in arbitrage activity. The results also reveal that the direction of prior arbitrage activity matters, and show that the decline in arbitrage intensity was especially pronounced for ETFs holding less liquid bonds, whose lead market makers had lower regulatory capital ratios, and for those associated with non-bank-affiliated APs. Finally, the paper provides a novel estimate of the elasticity of primary market activity to ETF premia, contributing to the literature on limits to arbitrage and intermediary frictions

High-frequency traders (HFTs) dominate secondary market trading in exchange-traded funds (ETFs) but do not engage in ETF arbitrage. By contrast, primary market arbitrageurs enforce the law of one price, but their activities are infrequent and limited by arbitrage costs. We find that primary market activity is associated with increased volatility and illiquidity in overweighted ETF constituent stocks, while HFT activity is linked to narrower bid–ask spreads. Using a quasi-natural experiment in Japan, we show that while ETF primary market activity can temporarily disrupt market quality, the liquidity benefits of secondary market trading ultimately outweigh these negative effects

Frequent, yet uninformed, market timing recommendations by a financial advisory firm generate significant flows for Chilean pension funds. These flows induce substantial changes in the Chilean foreign exchange rate due to the funds’ high allocation to international securities. Local banks provide liquidity to pension funds in the spot market and their hedging transactions propagate the demand fluctuations from the spot to the forward market, resulting in deviations from covered interest rate parity. Using bank balance sheet data, we confirm that banks’ risk bearing constraints create limits to arbitrage

We examine a link between bond exchange-traded fund (ETF) creation and redemption processes and the underlying bond market liquidity. Using daily creation and redemption data, we find that including a bond in a creation or redemption basket has a favorable impact on the bond’s liquidity for both high-yield and investment-grade markets. The improvement in liquidity persists during times of market stress with this impact being stronger for redemptions than creations. Our results suggest that ETF mispricing arbitrage explains the improvement in bond liquidity. However, we also find evidence that transaction costs and bond inventory management limit the ETF arbitrage

Short interest is currently required to be disclosed twice per month, but regulators have sought to increase this frequency. Meanwhile, short interest information from private third‐party vendors has emerged to meet investor demand on a daily basis. We find that daily private‐sector data strongly predict bimonthly regulatory disclosure. Furthermore, private‐sector data help price discovery, albeit with modest economic magnitude. Investors tend to underreact to the information content of private‐sector data mainly due to limits to arbitrage rather than market inattention. Despite the costly access to private‐sector data, we find no evidence that retail investors are harmed in their trades. Overall, our findings highlight the interplay between private‐sector and regulatory solutions in enhancing financial market transparency

This paper shows that leading theories of the beta anomaly fail to explain the anomaly’s conditional performance. Abnormal returns and Sharpe ratios of betting-against-beta (BAB) factors rise following months with below-median realized volatility, even controlling for mispricing, limits to arbitrage, lottery preferences, analyst disagreement, and sentiment. Moreover, the leverage constraints theory counterfactually predicts that market and BAB Sharpe ratios increase with volatility. We further show that institutional investors shift their demand from high- to low-beta stocks as volatility increases, and the resulting price impact is sufficient to explain the difference in abnormal BAB returns between high- and low-volatility states