Who actually shorts, and why is most of it not a bet against a company?

The public image of a short seller is a research fund publishing a report accusing a company of fraud. Those exist, and they are a tiny fraction of the borrow demand in any market. Most short selling is plumbing.

The real sources of demand

• Long/short and market-neutral equity. A fund holds $200m long in one basket and $200m short in another. Net market exposure is near zero; the position expresses a relative view about which basket does better. The short leg is as much a hedge as a bet.
• Pairs trading. Long one carmaker, short another. If the sector falls, both legs move together and the P&L is the spread between them. The short exists to remove the sector risk, not to attack the second company.
• Convertible arbitrage. Buy a convertible bond, then short a delta-weighted number of the underlying shares to strip out the equity exposure and isolate the bond's yield and volatility. A convertible exchangeable into 100,000 shares with a delta of 0.5 is hedged by shorting 50,000 shares, adjusted as the delta moves. Convertible arbitrage has historically been one of the largest single sources of borrow demand — created, note, by the issuing company's own financing decision.
• Index and ETF market making. Authorised participants and market makers routinely run short baskets or short ETF positions intraday as part of creation and redemption. This is liquidity provision with no directional content whatsoever.
• Merger arbitrage. In a stock-for-stock deal, the arbitrageur is long the target and short the acquirer in the exchange ratio. The short is how the spread is captured; it is not a view on the acquirer.
• Hedging a concentrated exposure. An investor who cannot sell a large holding — for tax, contractual or liquidity reasons — may short a correlated instrument to reduce the risk.
• Directional short research. The smallest slice by capital deployed, and by a wide margin the largest by share of headlines.

Price discovery, and what the research actually found

The economic argument for permitting short selling is that pessimists carry information too, and if they cannot trade, prices reflect only the optimists.

• Theory. Diamond and Verrecchia (1987) modelled exactly this: short-sale constraints slow the incorporation of negative information into prices.
• Cross-country evidence. Bris, Goetzmann and Zhu (2007), "Efficiency and the Bear", found that prices incorporate negative information faster in markets where short selling is permitted and practised.
• The 2008 bans. Boehmer, Jones and Zhang (2013) studied the US ban on short selling in roughly 800 financial stocks and found market quality deteriorated — wider spreads, less depth, more intraday volatility — without clear evidence of price support. Beber and Pagano (2013), covering some 30 countries during the 2007–09 crisis, found bans were associated with worse liquidity and slower price discovery, and were not generally associated with better share price performance. The then-chairman of the US securities regulator said publicly at the end of 2008 that, knowing what they knew by then, they would not have imposed the ban again.

The counter-case, stated fairly

This is a genuine controversy, not a settled one. Regulators facing a disorderly market argue that a temporary ban interrupts destabilising feedback loops, buys time for information to arrive, and protects against coordinated pressure on institutions whose funding depends on confidence. They also point out that academic measures of "market quality" — spreads, depth, volatility — are not the same as systemic outcomes, which are far harder to measure. Whatever the evidence on liquidity, short-selling bans remain a standard crisis tool and have been used repeatedly since 2008.

Both sides of that argument are worth understanding. This course takes neither.

Try it now

1. Take a company that funds itself with convertible bonds: Strategy (formerly MicroStrategy, MSTR.US), whose annual report for 2025 lists six series of convertible senior notes outstanding at the end of the year, about $8.2 billion in all. A balance sheet shows how much a company owes but not on what terms, so the conversion feature comes from the filings. The balance sheet is below, then the filings tab where the annual report sits.

Open MSTR.US — filings in the EODHD Terminal

2. Reason through the borrow demand its issuance created: bondholders hedging their delta must short the underlying shares, and must adjust as the share price moves.
3. The company's short interest is below, as a fraction of the float. Ask the diagnostic question: of that short interest, how much of it is a bet against the business, and how much is a hedge the company itself caused by choosing convertible financing? The honest answer is usually "you cannot tell from the short interest figure alone".