1929 and the Great Depression — what leverage actually does

The answer

Between September 1929 and July 1932 the Dow Jones Industrial Average fell from 381.17 to 41.22, a decline of about 89%. The crash itself was not the disaster. The disaster was that a large part of the buying had been done with borrowed money, so falling prices forced selling, which forced more selling, and took the banking system with it.

Why this costs you money

Leverage does not feel like risk. It feels like access.

Here is how it actually works on a person. You buy ₹5,00,000 of stock with ₹1,00,000 of your own money and ₹4,00,000 borrowed from your broker. The stock falls 20%. Your position is now worth ₹4,00,000. The loan is still ₹4,00,000. Your money is gone — all of it — on a 20% fall.

That is the arithmetic everyone knows. Here is the part that costs more.

At an 18% fall, before you are wiped out, your broker asks for more money. This is a margin call: a demand to add cash or the position is sold. You have, typically, hours. Not days.

So the decision about whether to hold through a fall is taken away from you at exactly the moment when holding is most likely to be right. You do not sell because you decided to sell. You sell because you were asked for money you did not have on a Tuesday afternoon.

Leverage does not just amplify losses. It removes your ability to be patient, and patience was the only real advantage you had over a professional.

How it works

In the 1920s an American investor could buy shares on margin with a down payment that was often 10% to 20% of the price, and sometimes less. Brokers borrowed the rest from banks, and by 1929 broker loans outstanding exceeded $8 billion.

Follow the chain, because the chain is the lesson.

1. Investors borrow from brokers to buy shares.
2. Brokers borrow from banks, using the shares as collateral.
3. Banks lend more as prices rise, because the collateral is worth more.
4. Prices rise partly because of the lending in step 3.

Steps 3 and 4 form a loop that runs both ways. On the way down it runs faster, for a reason worth stating precisely: buying is optional, selling under a margin call is not. A rise is made of willing buyers. A fall is made of forced sellers, and forced sellers do not care about the price.

The dates, for the record.

• 3 September 1929 — Dow peaks at 381.17.
• 24 October 1929, "Black Thursday" — heavy selling, a fall of about 11% during the day, partly recovered by the close.
• 28 October 1929, "Black Monday" — down 12.8%.
• 29 October 1929, "Black Tuesday" — down 11.7%.
• 8 July 1932 — Dow closes at 41.22, the low.

Then the second stage, which made it a depression rather than a crash. Banks that had lent against shares held collateral worth a fraction of the loans. Depositors, hearing of losses, withdrew money. Banks that were solvent but not liquid failed anyway. Estimates of US bank failures between 1930 and 1933 are commonly given as around 9,000. Unemployment reached about 25% in 1933.

The Dow did not regain its 1929 level until 1954. That is 25 years. "Just hold on" is true over long enough periods, and it is not free.

What it tells you, and what it does not

What it tells you. Leverage converts a price decline into a liquidity event. The question stops being "is this company still good?" and becomes "can I find cash by 3 p.m.?" Those questions have different answers and only one of them is about investing.

It also tells you that leverage is often invisible in aggregate. In 1929 nobody could see the total. Each broker knew their own book. Each bank knew its own loans. The system-wide number existed only after the fact.

What it does not tell you. It does not tell you that borrowing to invest is always wrong. A home loan is leverage. A business borrowing to build a factory is leverage. Leverage against an asset with stable cash flows and a long repayment schedule is a different animal from leverage against a share price with a same-day margin call.

The distinction that matters is not "debt or no debt". It is who can demand the money back, and how quickly. A 20-year home loan cannot be called because the property fell in value. A margin loan can be called this afternoon.

And 1929 does not tell you that crashes cause depressions. Economists have argued for decades about the causal chain. Friedman and Schwartz put the weight on monetary policy: the Federal Reserve allowed the money supply to contract by about a third between 1929 and 1933. Others emphasise the gold standard, trade policy including the Smoot-Hawley tariff of 1930, and the absence of deposit insurance. The crash was one part of a system that had several failures at once.

The decision rule

Before any position, ask: what price forces me to sell? If such a price exists, that price is your real risk, not the company's prospects.

If the answer is "no price forces me, I own this outright", you have the freedom to be wrong for 3 years. That freedom is worth more than most strategies.

If the answer names a price, then write it down and treat it as the position's true stop, whatever your intentions are. The conditional version: leverage is acceptable when the loan has a fixed term you can survive and the lender cannot accelerate it on price alone. It is dangerous when the lender can demand cash the same day, because your good judgement will not be consulted.

Try this now

Five minutes. Most readers will start by thinking the answer is zero, and then remember something.

1. Open your broker app and find the section called Margin, MTF, Margin Trading Facility, e-margin, Pay Later, or in the United States, Margin balance or Margin debit. Write down the amount owed.
2. Open your holdings and look for any position marked as pledged. In India, check the CDSL or NSDL statement that is emailed to you monthly, and look for shares pledged in your own name. Pledged shares are usually collateral for a loan.
3. Check for a loan against securities or loan against shares from a bank or NBFC.
4. Now the part people forget. Ask 3 questions.

• Did any money invested in the last 3 years come from a personal loan, a gold loan, a top-up on a home loan, or a credit card?
• Do you hold any futures or options position? A futures contract is leverage even though no loan appears anywhere. Find the contract value, not the margin you paid.
• Are you paying an EMI on anything at all while also holding shares? If yes, part of your portfolio is financed by that debt, whatever the paperwork says.

1. Add it all up. Divide by the total value of your investments.

What you should see. A single number: the percentage of your portfolio that is borrowed. For most readers it will be 0% and that is a good answer. For many it will be between 5% and 30%, which is usually a surprise, because it arrived in pieces and no screen ever adds it up.

Then do the arithmetic that 1929 forces. Take your borrowed percentage. Work out the fall in your portfolio value that would wipe out your own money. If you are 30% borrowed, a 30% fall takes everything you put in. Indian and US markets have both fallen more than 30% several times in living memory.

Three real cases

1. Broker loans in the United States, 1929 — the system-wide number nobody could see Outstanding broker loans exceeded $8 billion by 1929, against a total market value of listed shares of roughly $87 billion. What it turned out to be: enough forced selling to overwhelm every willing buyer for 3 years. After the crash, the Securities Exchange Act of 1934 gave the Federal Reserve power over margin requirements, exercised through Regulation T, which has required 50% initial margin for decades.

2. Long-Term Capital Management, 1998 (United States) — leverage without stupidity A fund run by extremely accomplished people, including 2 Nobel laureates in economics. Its strategies were mathematically careful. Its leverage was roughly 25 times its capital. When Russia defaulted in August 1998, the positions moved against it and the size of the positions meant it could not exit without moving prices further against itself. The Federal Reserve Bank of New York organised a rescue by a group of banks in September 1998. What it turned out to be: proof that leverage is not a failure of intelligence. It is a failure of survivability.

3. Indian promoter pledging, 2018 onwards — the same loop, listed today Several Indian promoters pledged shares in their own companies to raise loans. When those share prices fell, lenders invoked the pledges and sold. The selling pushed prices lower, which triggered further invocation. Companies in infrastructure, media and financial services were affected in 2018 and 2019 . What it turned out to be: the 1929 chain, running inside individual companies, disclosed quarterly to the exchanges, and generally ignored by shareholders until the selling started.

The question that resolves it

A novice asks: how much can I make on this?

An expert asks: what happens to me if this falls 40% next month?

The second question has an arithmetic answer, and if you have leverage, the answer is not "I hold on and wait". The answer is "I am sold out at the bottom by somebody else".

What would make this wrong

The falsification condition. If leverage merely amplified outcomes symmetrically, then leveraged investors would show the same win rate as unleveraged investors with larger swings in both directions. They do not. The asymmetry comes from forced exit: a leveraged investor can be removed from a position that later recovers. If a study showed leveraged retail investors achieving the same long-run returns as unleveraged ones, scaled up, the argument here would be wrong.

The honest limit. This article is about leverage that can be called. It is not an argument against all borrowing. Somebody with a stable income, a 15-year loan at a fixed rate, and no clause allowing the lender to demand repayment early, is running a genuinely different risk. Treating those 2 cases as the same thing is lazy.

The hindsight trap. It is easy to say that 1929 was an obvious bubble. Look at what a careful American investor actually knew in the summer of 1929.

The 1920s boom was not imaginary. Electrification, the motor car, radio and mass production were transforming output. Corporate profits had genuinely risen through the decade. Consumer prices were stable, so this was not a currency panic. The Federal Reserve, created in 1913, was widely believed to have made banking panics a thing of the past. And Irving Fisher of Yale, one of the finest economists of the era, said in October 1929 that stock prices had reached "what looks like a permanently high plateau".

So what could a careful person genuinely have seen? Three things, all published at the time.

First, the level of broker loans. It was reported, and it had roughly doubled in 2 years. Anybody could see the buying was increasingly financed.

Second, the behaviour of investment trusts, the closed-end funds of the day. Many held shares in other investment trusts, each layer adding leverage, and some traded far above the value of what they owned. That gap was calculable from public data.

Third, the real economy had already turned. Industrial production peaked around mid-1929, before the crash. A careful person watching output rather than prices had a signal months early.

What that person could not have seen was the depth or the length. Almost nobody predicted an 89% fall or 25 years to recover. The bears of 1929 who are famous today were, on average, early by a year or more, and several bought back in 1930 and lost anyway. Knowing that a market is stretched tells you how to size a position. It does not tell you what to do on Monday.

Carry this

• Ask of every position: what price forces me to sell?
• Exposure ÷ your own money. Above 1, you are leveraged, whatever it is called.
• A fall of X% wipes out your capital if you are X% borrowed. Both markets have done 30% more than once.

Knowledge check

Related

• Black Monday 1987 and the machines
• The 2008 global financial crisis
• The anatomy of a bubble — what they all share
• ← Market history: manias, crashes and manipulation