The Campbell–Shiller log-linear approximation is not a universally valid accounting identity, because its interpretation and validity depend on restrictive assumptions about dividends, investor beliefs, and pricing models that often fail in real-world data, from Itzhak Ben-David and @alexchinco https://t.co/l27ERKSPzC
Remember how ETFs can rebalance without ever incurring capital-gains tax?
So what if you seed ETFs with appreciated stocks?
My story today on the latest idea from Wall Street's booming tax-alpha complex: https://t.co/2Ae8DVCj1T
@profplum99 Not arguing that it’s the right thing to do, but comparing EY to Rf makes a lot of sense in a world where people are focused on maximizing EPS.
https://t.co/xZZgpOv2Sk
@corry_wang Completely agree! We’re trying to bring this to researchers’ attention. Analysts put most of their time into E[EPS]. Then they ask themselves: “How has the market priced each $1 of the company’s earnings in the past?” Here’s how we put it on p5 in the intro:
@wagonomics@arpitrage Analysts spend most of their time on E[EPS]. Our point is that they convert E[EPS] into a price target by asking themselves: “How has the market priced each $1 of this company’s earnings in the past?” Regardless of how accurate E[Price] is, there’s no discount rate involved.
Lots of research has gone into figuring out how sell side analysis (ie in IBES) form expectations. But if you read the reports, it turns out they forecast earnings per share and slap on a trailing P/E multiple
By @AlexChinco and Ben-David
https://t.co/IX0WbWRAit