Leverage

011 min

Leverage at a glance

• What it is: A fixed input — capital, labor, code, or media — producing output far larger than the input itself.
• The number that matters: Lehman Brothers' leverage ratio hit 30.7 to 1 by fiscal 2007, months before its 2008 bankruptcy.
• Where it breaks: The same ratio that multiplies a gain multiplies a loss by an identical amount.
• Where it comes from: Named for Archimedes' lever — a small force, placed right, moves a weight many times its size.

021 min

The problem Leverage names

Teams and investors often size a decision by the resource it visibly consumes — the cash spent, the hours booked — while missing the multiplier already built into the arrangement. A founder who borrows to fund inventory is not simply spending money: every dollar of debt sits on top of a smaller amount of equity, so a swing in the business's performance changes equity by a much larger percentage than the raw numbers on the balance sheet suggest.

The same blind spot shows up outside finance. A twelve-person team that ships software running for a hundred million users has not done twelve people's worth of work; the code itself keeps working for every new user, at no added cost per user, long after the twelve people who wrote it moved on to something else.

Missing this costs in both directions. A company that ignores its own leverage underprices real risk — a modest revenue dip becomes an equity wipeout nobody modeled for. A company that ignores a competitor's leverage prices its own response too high — matching headcount for headcount against a rival whose product adds the next user at close to zero marginal cost.

032 min

Understanding Leverage

The word describes a physical fact before it describes a financial one. A lever multiplies force: push down with a small amount of force far from the pivot, and the other end moves a much heavier load close to the pivot, because the ratio of the two forces scales with the ratio of the two arm lengths. Business leverage is the same ratio without the physical bar: whatever fixed input a person or company puts in — borrowed capital, other people's time, a piece of software, a piece of media — some feature of the arrangement lets a small change in that input produce a large change in the output.

In finance the mechanism is arithmetic. A company's leverage ratio is its total assets divided by its shareholders' equity. A firm with no debt in its capital structure has no financial leverage at all, and its equity is what OpenStax's Principles of Finance calls unlevered equity. Add debt without adding equity, and the ratio rises: the same equity base now stands behind a larger pool of assets. If those assets earn more than the interest rate on the debt, the surplus flows entirely to equity, and the smaller the equity base, the larger its percentage gain. The identical mechanism runs in reverse: a loss on the assets is absorbed by the same small equity base first, so the percentage loss to equity equals the percentage loss on assets multiplied by the leverage ratio. Debt does not choose a direction; it multiplies whichever direction the underlying assets move.

Labor, code, and media follow the same shape without a balance sheet. A manager directing ten people is multiplying their own judgment by ten sets of hands. A program that runs for a million users is multiplying one developer's decisions by a million executions, at no added cost per execution. In every case, some part of the arrangement — debt, headcount, or a copy that costs nothing to make — turns one unit of input into many units of output, and that multiplier is what the word names.

041 min

Where Leverage comes from

The mechanical sense of the word predates the business one by roughly two thousand years. The earliest surviving version of the claim comes from Plutarch's Life of Marcellus (c. 45–120 CE), which records Archimedes telling King Hiero of Syracuse, in the Doric dialect: "Give me a place to stand and with a lever I will move the whole world." The boast followed from Archimedes' own mathematics of the lever, which showed why a small force at a long distance from the pivot balances a large force close to it.

The business sense is much younger and has no single named originator. "Leverage" entered finance simply as the natural word for using debt the way a lever uses distance — a small equity stake controlling a much larger asset base. The broader reading, that labor, code, and media follow the identical multiplying logic, was named explicitly by investor Naval Ravikant in a 2018 Twitter thread, "How to Get Rich (without getting lucky)," later expanded in a recorded conversation published on his own site.

051 min

Types of Leverage

Ravikant's framework sorts leverage into four types, distinguished by what stands behind the multiplier.

Labor is the oldest form: other people's work. Ravikant puts it plainly: "The oldest form of leverage is labor, which is people working for you." A manager who directs ten people multiplies their own judgment by ten pairs of hands, and society, he argues, "overvalues labor as a form of leverage" precisely because it is the easiest kind to see.

Capital is other people's money, including debt. It multiplies a fixed equity stake into control over a much larger pool of assets — the mechanism behind Lehman Brothers' leverage ratio, described below.

Code is software that keeps running once written, at close to no cost per additional user. Ravikant's own example: "one great engineer can for example create bitcoin, and create billions of dollars worth of value," work that a payroll headcount could never explain.

Media is writing, video, or audio that keeps reaching people after it is published, without the creator's further time. In Ravikant's words, "coding, writing books, recording podcasts, tweeting, YouTubing, these kinds of things, these are permissionless" — anyone willing to make the thing can use this form, unlike labor and capital, which someone else has to grant.

081 min

What this means for your work

A product manager comparing two roadmap bets can ask which one is code leverage and which is a one-time service: a self-serve onboarding flow that a customer completes without a salesperson keeps paying out to every future customer at no added cost, while a bespoke integration a solutions engineer rebuilds for each account never compounds past the account it was built for. The PM who tracks that difference funds the first bet even when it costs more engineering time upfront.

An engineer choosing between a one-off script and a reusable internal tool is making the same call at smaller scale: the script solves today's ticket, while the tool solves every future ticket like it for the same one-time cost, provided a colleague can find and run it without asking the person who wrote it.

A founder raising a venture debt facility instead of selling more equity is taking on financial leverage directly: the debt has to be repaid on a schedule regardless of next year's revenue, so the decision should be sized against the company's worst plausible year, not its average one.

A designer who ships one well-made component to a shared library, instead of a one-off screen, is choosing code leverage over a service: every team that adopts the component afterward gets the same quality for zero extra design hours.

091 min

How to apply Leverage

1. 1

   Name the fixed input and the claim behind it.

   Debt and rent have to be paid on schedule regardless of results; a codebase carries no such claim, which is why the two forms behave differently on the way down.

2. 2

   Size a financial position against the worst plausible year, not the expected one.

   A leverage ratio that looks safe against an average outcome can erase equity against a bad one, because the ratio multiplies a loss by the same factor it multiplies a gain.

3. 3

   Measure judgment separately from headcount when using labor leverage.

   Directing ten people through a bad decision multiplies that decision by ten; the multiplier has no opinion about which direction it multiplies.

4. 4

   Check the real marginal cost of the next unit for code or media leverage.

   If serving the next customer or reaching the next reader costs meaningfully more than the last one did, the leverage is smaller than the headline number suggests.

5. 5

   Recheck the ratio, not the absolute size.

   A company that has grown its assets and its equity by the same percentage has not added leverage, even though every number on the page got bigger.

101 min

When Leverage breaks

Leverage does not stop working when the underlying result is a loss; it keeps working, multiplying the loss by the same factor that would have multiplied a gain. The mechanism does not distinguish between the two directions.

The clearest documented case is the run-up to 2008. By fiscal 2007 Lehman Brothers carried a leverage ratio of 30.7 to 1, and Bear Stearns's ratio had, according to contemporary reporting, "rose sharply, to 33 to 1." A decline in asset values of only a few percent was enough to erase shareholders' equity entirely. When mortgage-backed assets lost value in 2008, the same ratio that had multiplied the preceding years' gains multiplied the losses, and Lehman filed for bankruptcy that September.

A second, quieter failure mode is leverage without a plan for the downside case. A company that borrows to expand inventory ahead of a busy season is using financial leverage correctly if the season arrives; the same debt becomes the reason the company cannot survive a season that does not arrive. The tell is a decision justified only by the expected outcome, with no separate answer for what happens if the fixed obligation — the debt payment, the salaries, the rent — still has to be paid and the expected outcome does not show up.

111 min

Leverage vs. nearby concepts

The confusion this audience makes most often is with Force Multiplier. Leverage names the ratio itself — how much output a fixed input controls — while a force multiplier is any specific factor, such as training or better tools, that raises that ratio. Leverage is the measurement; a force multiplier is one way to improve it.

A second, unrelated confusion is with Leverage Points, a systems-thinking term for the places in a system where a small intervention produces a large system-wide change. It uses the same physical image but names something different: a leverage point is a location in a system, not a ratio between an input and an output.

122 min

Where the Leverage evidence is contested

That leverage amplified the losses in the 2008 financial crisis is not disputed. What is disputed is a specific claim about why bank leverage rose in the first place.

In August 2008, former SEC official Lee Pickard argued in the American Banker that a 2004 change to the SEC's net capital rule let broker-dealers raise their leverage past a traditional 12-to-1 limit. The New York Times repeated the claim that October, reporting that "at Bear Stearns, the leverage ratio ... rose sharply, to 33 to 1." A law professor and a Nobel laureate economist both cited the same story afterward.

Economists Andrew Lo and Mark Mueller disputed it, using the SEC's own record. The SEC's director of market regulation, Erik Sirri, had already stated in an April 2009 speech that "the Commission did not undo any leverage restrictions in 2004." A 1999 U.S. Government Accountability Office review had already found that three of four broker-dealer holding companies "had ratios equal to or greater than 28-to-1 at fiscal year-end 1998, which was higher than their ratios at fiscal year-end 2006 before the crisis began" — years before the 2004 rule, using numbers that had been sitting in public SEC filings the whole time. Lo and Mueller's own reading: that trained economists and regulators could all repeat the same checkable error shows that "sophisticated and informed individuals can be so easily misled on a relatively simple and empirically verifiable issue."

Where this leaves a practitioner: the high leverage ratios were real, and they did precede the losses. Exactly which regulatory change, if any, caused banks to seek that leverage is the part still argued over.

132 min

Frequently asked questions about Leverage

141 min

Related Concepts

• Force Multiplier — a specific factor that raises the leverage ratio, not the ratio itself.
• Leverage Points — the systems-thinking term for where to intervene, not how much a fixed input controls.
• Diminishing Returns — the opposite pressure: why one more unit of input eventually buys less output.
• Opportunity Cost — what a decision that uses leverage gives up by committing a fixed resource one way.
• Compounding — leverage multiplies an outcome once; compounding multiplies it again each period.

?6 questions

Questions people ask

§8 sources

Sources