Interpreting FINRA Reg SHO Short-Volume Data

# What does Short Volume in FINRA RegSho actually measure?


# What does Short Volume in FINRA RegSho actually measure?












When reading RegSho, I see a table like this:

```
Date        Symbol   ShortVolume    ShortExemptVolume   TotalVolume    Market
20210202    A        93137          2300                297479         B,Q,N
20210202    AA       1174892        2077                3055114        B,Q,N
20210202    AAA      8              0                   8              Q,N
20210202    AAAU     45332          0                   107340         Q,N
20210202    AAC/U    459206         42465               6146462        Q,N
...
20210202    GME      16358136       1073011             29733410       B,Q,N
```

What do these numbers actually represent? People on reddit are claiming that ShortVolume is the sum of shares in transactions where the seller did not actually have shares to sell, and that they must buy the shares back within 3 days. I see similar claims on nakedshortreport, but as an amateur trader, I don't know if either is a trustworthy source. If that explanation is correct, I believe that would imply that over the past 5 days, short-term IOUs have been generated for 60% of GME's total market cap.

Are these explanations of the data correct? If so, is the data limited in some way that an amateur trader might be unaware of?

## Answer by Tim C (score 0)

https://quant.stackexchange.com/a/60923

I've done a bit more research and I'll attempt a self-answer here.

#### The top level definition is correct, all of the inferences that follow it are wrong.

> What do these numbers actually represent? People on reddit are claiming that ShortVolume is the sum of shares in transactions where the seller did not actually have shares to sell

Up to this point, the assumptions in the question are correct.

#### Wrong Inference 1: "A short sale must be closed by a buy within 3 days"

> ...and that they must buy the shares back within 3 days

No; the seller could have an agreement to borrow the shares. They do not need to buy them on the open market. With such an agreement, the short seller could have entered a short position, which they can hold indefinitely subject to the terms of their borrowing agreement.

#### Wrong Inference 2: "Only non-short transactions can close short transactions"

> If that explanation is correct, I believe that would imply that over the past 5 days, short-term IOUs have been generated for 60% of GME's total market cap.

This is also not true. Unpacking the math here a little bit - the 60% number came from the sum of short transactions over 5 days, minus the sum of non-short transactions over the same period (and then dividing it by the number of GME shares issued). It relies on the implicit assumption that a short volume over 50% means that a short position has increased, which is not true.

Short transactions can close other short transactions: if someone buys back their own IOU, that IOU can be closed without any non-short transactions taking place. Additionally, the FINRA data does not include trades that are not publicly disseminated; it is possible that a trader in the short transaction settled in a trade that did not happen on the open market.