Understanding the Mechanics – and Metrics – of Short Selling

Short selling has existed in its most basic form for hundreds of years, but the practice of taking a position that profits from the decline of an asset’s price has become rather sophisticated in modern times – as well as controversial. We thought it would be useful to put together an educational overview so that investors can better understand the process as well as the related metrics.

The mechanics of short selling

At a fundamental level, selling short is the process of borrowing a security, selling it at a higher price with the expectation that the price will fall, and then subsequently buying back the security and returning it to the lender to close out the position. In other words, the process inverts the traditional “buy low, sell high” adage to become “sell high, buy low.” 

Once a short seller has borrowed shares and sold them in the open market, the investor still must return the borrowed stock at a later time to close out the position and realize any gain or loss. This is why open short positions can be seen as pent-up buying demand. While a short seller can potentially hold the short position for an extended period of time, the market’s natural tendency to trend higher over long time horizons adds potential pressure for a short seller to close out the position.

In contrast, long investors can hold on to stock indefinitely with little to no structural pressure to sell the shares. Furthermore, there are other mechanics related to short selling that make maintaining short positions more complicated than holding long positions, primarily related to the fact that stock must be borrowed before being sold short. At the same time, the securities lending industry that facilitates short transactions has become quite sophisticated and interconnected.

The metrics of short selling

There are many metrics that are relevant to understanding short selling activity. The first is short interest, which is simply the number of shares that are being held in short positions by short sellers. US exchanges report official short interest figures twice per month, but the metrics are released approximately 7 business days after the associated settlement date. That delay makes the data less actionable, but the exchange reported data is the most official source for short interest information.

Days to cover measures how many days it would hypothetically take for all short positions to be covered based on a stock’s average volume. Shares on loan represents how many shares have been borrowed and remain on loan. It’s important to note that shares that have been loaned may not have necessarily been sold short. Utilization shows how many shares that are available from securities lenders are currently on loan. Again, keep in mind that shares that have been lent does not necessarily mean that those shares have been sold short (more on this later).

Cost to borrow is the fee that a short seller must pay to borrow stock, and is presented as an annualized percentage rate. Due to the nature of the securities lending market, these fees can vary dramatically depending on a wide range of factors, including supply of borrowed shares as well as demand from short sellers. If a short seller has a high conviction bearish thesis, they may be willing to pay higher fees in order to execute the short trade. A high utilization rate does not necessarily mean that short sellers are willing to pay high borrowing fees.

Here is a list of the most relevant metrics that we display on ORTEX.

  • Shares Outstanding: The total number of shares that a company has issued.
  • Free Float (FF): The number of shares that are available for trading when excluding shares that are restricted from trading or closely-held. For example, company insiders will often hold large positions but are restricted from selling. Long-term investors such as institutions or index funds may also hold stock that the entity has no intention to sell.
  • Short Interest (SI): The number of shares that are being held in short positions by short sellers.
  • Exchange Reported SI: The official number for short interest provided by the exchanges. However, it is only reported in the US twice per month based on settlement date and the release is delayed by several business days. Additionally, reporting frequency and delays differ by region and market.
  • Exchange Reported SI % of FF: The official number for short interest presented as a percentage of free float.
  • ORTEX Estimated SI: Our proprietary real-time intraday estimate of current short interest, which is derived by using exchange SI as reference points and making adjustments based on stock lending data and other data sources.
  • ORTEX Estimated SI % of FF: Our proprietary real-time intraday estimate of current short interest presented as a percentage of free float.
  • Days to Cover (DTC): The number of days it would take for all short interest to be covered, based on average volume.
    • ORTEX provides additional ways to measure DTC by using average volumes over different time frames. We also calculate DTC for shares on loan for those additional time periods.
  • On Loan: The current number of shares that are out on loan. Stock that is on loan has not necessarily been sold short.
  • On Loan – Average Age: The weighted average number of days that current stock loans have been outstanding. 
  • On Loan – Number of Loans: The number of stock loans that are outstanding on a particular day.
  • Utilization: The percentage of shares that are available at lending programs in the wholesale market that are currently outstanding on loan.
  • Cost to Borrow (CTB): The contractual fee that a borrower has agreed to pay a stock lender. This figure is presented as an annualized interest rate. Fees vary across contracts, and ORTEX data includes the minimum, maximum, average, and standard deviation of all loans, new loans, and returned loans on a given day.
    • ORTEX breaks down CTB into various categories. We provide average, minimum, maximum, and standard deviation CTB figures for all loans, new loans, and returned loans.
  • Failure to Deliver (FTD): The total number of shares that have failed to be delivered to settle trades. There are many reasons that contribute to FTDs and FTDs are not necessarily indicative of “naked” short selling (shorting stock that has not been borrowed).
  • Threshold Securities: Stocks that have FTDs in excess of 10,000 shares or represent at least 0.5% of total shares outstanding for five or more consecutive settlement days.

While many of these metrics are interrelated in various ways, it’s important to understand that there are countless other factors and variables at play in the dynamic securities lending market that is full of millions of market participants. At ORTEX, we strive to offer investors the most comprehensive view into the securities lending market, leveraging our extensive array of data sources and proprietary algorithms to provide actionable insights.




Where do loaned securities come from?

There are several prominent sources of shares that are available to borrow.

The largest source of securities for lending comes from large institutional investors such as pension funds, mutual funds, and insurance companies. 

Another source is related to margin loan hypothecation, which is where an investor has an outstanding margin loan balance with his/her/their broker-dealer. As collateral for that loan, the broker-dealer can secure some of the underlying securities that were purchased and choose to lend those shares out to short sellers. If that investor chooses to pay off that loan – either by selling securities or depositing cash – then the broker-dealer no longer has access to those shares. In that situation, the broker-dealer may recall the shares and close out the short seller’s position simply because the borrowed shares are no longer available.

Borrowing securities usually entails a fee, typically expressed as an annualized percentage rate, that will fluctuate based on market conditions. Institutional or retail investors that own shares outright (i.e. there is no outstanding debt associated with the position) can choose to voluntarily lend their stock to short sellers in exchange for the fee compensation. That income can boost returns, while long-term investors may not be particularly concerned with any short-term volatility attributable to short selling activity. 

Borrowing costs can fluctuate significantly based on a wide range of factors, including the supply and demand for borrowed shares, collateral flexibility, and the term of the transaction, among other things. Fees are often privately negotiated between investors and lenders, and a short seller with a strong conviction to initiate a bearish strategy may be willing to pay a higher fee.

Securities lenders can also put securities on hold for an investor in anticipation of the trader initiating a short position, which is known as “icing.” The lender does not initially receive a fee for this service, but another interested borrower can challenge the hold by placing a bid for the available shares. In that situation, the initial investor usually has approximately 30 minutes to choose between utilizing the shares or releasing them. 

Are there other reasons to borrow stock?

Beyond short selling, there are several reasons why investors seek to borrow stock. The most prominent reasons are:

  • Settlement certainty: When an investor borrows a share to avoid a failed delivery of a separate trade.
  • Market-making: When a broker lends a client’s sell list and borrows the buy list, netting out positions over time to avoid impacting the market.
  • Collateral: Using borrowed securities such as stocks or bonds as collateral for another trade, such as a swap or other derivatives transaction.
  • Dividend arbitrage: Lending out a stock to a domestic investor over the dividend record date to avoid foreign investor withholding taxes.

How can short interest exceed 100% of free float?

In recent years, investors have taken note of a particular phenomenon related to short selling. Certain stocks have at times seen short interest that is greater than 100% of free float. That may initially seem impossible, but here is how it works.

Consider investor A holds 100 shares, and chooses to lend those shares to investor B. Next, B then short sells the 100 shares to investor C, who is now long 100 shares. If investor C then lends out the 100 shares to investor D to sell short, then there are now 200 shares held short in this example after starting with 100 shares long. This shows how there can be a multiplier effect on tradable shares as the lending cycle continues. 

The process is similar to how a traditional bank operates and influences money supply. A customer deposits cash, which the bank will then use to fund loans, which subsequently increases the money supply circulating in the economy.

That is how enough short selling and securities lending activity can cause short interest to exceed 100% of free float.

How is ORTEX data more accurate than other analytics platforms?

ORTEX provides real-time estimates for short interest that is derived from all of the metrics we measure. We use official exchange data and calculate our estimates based on intra-day stock lending activity, as we receive daily stock lending data from approximately 85% – the largest pool of sources – of all securities lending sources. 

Since it is impossible to know short interest at any moment with 100% certainty, we also provide confidence intervals for our estimates that may fluctuate depending on market conditions such as volatility. Our live data provides intra-day insights as securities are dynamically lent and returned throughout the day.