Aleksandra Rzeźnik

In statistics, samples are drawn from a population in a data-generating process (DGP). Standard errors measure the uncertainty in estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence-generating process (EGP). We claim that EGP variation across researchers adds uncertainty—nonstandard errors (NSEs). We study NSEs by letting 164 teams test the same hypotheses on the same data. NSEs turn out to be sizable, but smaller for more reproducible or higher rated research. Adding peer-review stages reduces NSEs. We further find that this type of uncertainty is underestimated by participants.

We study the link between illiquidity and co-movement in illiquidity and the way asset managers trade off illiquidity and co-illiquidity in their portfolio allocation decision. By exploring two experiments – the 2005 SHO Regulation and 2016 Tick Size pilot program – we document the way fund managers manage co-illiquidity risk and the implication for the market degree of illiquidity and co-illiquidity.

Active mutual funds supply liquidity when demanding it becomes uneconomical. They tilt toward cheaper buy trades after inflows deplete their trading ideas, when trading ideas in general run low, and when they have more stocks to supply liquidity to, and their cheaper trades perform worse. Their largest trades are more likely to supply liquidity, explaining why they were not broken up. Funds perform better when they pay more for their buys, and perform worse when they pay more for their sells, consistent with the implied value of the trades and the correlation between what a fund trades and what it holds.