Part III · Chapter 7 · Leverage · lesson 6

When leverage is the wrong tool

10 min

Builds on Margin, mortgages and the forced sale

A payment schedule is a promise about specific amounts on specific dates. It does not know what kind of year you are having.

Almost every leverage failure is that sentence, applied to somebody whose income, asset or timetable had a different shape from the promise they made against it. Four shapes, and each one has a name.

Leverage does not create a return. It multiplies a return that already exists, and charges interest for doing so.

Which means the tool is wrong whenever the thing being multiplied is not reliably there — and wrong in a specific way, because the cost is reliably there regardless.

One: income that moves

A fixed obligation serviced by a variable income is a mismatch of shapes, and the mismatch is worst when the two are connected.

A salaried employee in a stable sector has income that varies little and varies independently of most asset prices. A commission earner, a contractor, a business owner or anyone paid out of the same industry the asset sits in has income that moves, and moves with the thing they borrowed against. Their bad month is the asset's bad month is the lender's nervous month.

The test is not whether the income is large. It is how far it can fall, and whether it falls at the same time as everything else.

Not the same thing

The obligation

What you have promised to pay, on which dates, under what conditions.

Measured in: An amount and a date, both fixed in advance
For example: A payment due on the first of every month for twenty-five years.
Chase it alone and: It is completely indifferent to your circumstances, which is exactly what makes it cheap — the certainty is what the lender is buying.

The income servicing it

What actually arrives, and how reliably.

Measured in: An amount, a variability, and a correlation with everything else
For example: Commission that was excellent last year and is a question this year.
Chase it alone and: It is assessed at its recent level rather than at its floor, and the floor is the only number the obligation will ever meet.

Which is which? Put each one on a side.

A repayment that rises when the reference rate rises
Rental income from a tenant on a one-year term
A bonus that has been paid every year for six years
Interest that is added to the balance when it is not paid in cash

Two: collateral that cannot be sold in pieces

Borrowing against something liquid gives you an option: sell part of it, reduce the debt, keep the rest. Borrowing against something illiquid gives you no such thing. The exit is all of it or none of it, it takes months, and the price is whatever a small number of possible buyers decide during those months.

That turns a manageable problem into an unmanageable one. A holder of liquid collateral facing a shortfall sells a tenth. A holder of illiquid collateral facing the same shortfall sells the whole thing or nothing at all, and if the deadline is short, the "whole thing" is sold at the price that clears in the time available.

The two costs from earlier in this course meet here: illiquidity is a discount you pay when forced, and leverage is the mechanism that does the forcing.

Three: a horizon shorter than the asset's bad patch

Every asset has a characteristic period over which it goes wrong. For some things it is weeks; for property and private businesses it can be years.

If the date you need the money is inside that period, the leverage does not amplify a return, it amplifies a coin toss about timing. The unleveraged holder with a short horizon takes the price on the day and gets on with their life. The leveraged one with a short horizon may not reach the day.

The practical version of this test is uncomfortable and short: name the date you need the money, and name the longest stretch this asset has spent below today's price. If the second is longer than the first, the horizon is the problem, and no amount of being right about the asset fixes it.

Predict

Someone borrows to buy an asset because they cannot afford the position otherwise, and the position is the only way to reach a number they need by a fixed date. What is the structural problem?

They cannot afford it, so they should not do itThe leverage is being used to close a gap rather than to multiply something that worksThe fixed date means they will pay tax on the gainBorrowing costs more than the asset will return

Four: not being able to state the trigger

The shortest test of all, and it subsumes several of the others.

If you cannot say, in one sentence, what price move ends this position and who decides — you do not yet know what you are holding. Not because the number is necessarily bad, but because a position whose failure condition is unexamined has not been sized against anything.

Something people say

“Debt is always a mistake.”

Why it spreads: It is excellent advice for the debt most people actually encounter: unsecured borrowing at high rates against consumption that produces nothing. Applied there, the rule is close to unimprovable, and the people who repeat it have usually watched it save somebody.
The part that is true: High-rate borrowing against a depreciating purchase has a negative hurdle that no asset return can clear, so the strong form of the rule is correct in the case that causes the most damage.

What the rule cannot distinguish is a fixed obligation matched to a stable income, against a durable asset, over a horizon longer than the asset's bad patch, at a rate below what the asset produces. That position and the credit-card balance are the same word and not the same object. Refusing the second is sensible; refusing the first has a cost, and the cost is usually paid in decades of rent or in a business that stayed small.

Where it is the right tool

To be even-handed about it, the conditions where leverage does its job are as statable as the conditions where it does not.

The obligation is small relative to an income that is stable and uncorrelated with the asset. The collateral can be sold in pieces, or the term is long enough that no sale is required. The horizon comfortably exceeds the asset's characteristic bad patch. The asset's return clears the hurdle with room. And the trigger price is known, written down, and far enough away that ordinary movement does not reach it.

That is a demanding list, and a great many sensible borrowings satisfy all of it. The point of the list is not to discourage — it is that each item is checkable in advance, which is the one moment when checking is free.

Check

Which of these positions is the most fragile, on the four tests in this lesson?

A long fixed-rate loan on a home, held by a salaried worker who intends to stay twenty yearsA short-term loan against a stake in a private business, held by someone paid out of the same industryA modest borrowing against liquid holdings by someone with large cash reservesA business borrowing against invoices already issued to creditworthy customers

Worth remembering

Leverage multiplies a return that already exists and charges interest whether or not the return appears.
The common failure is a shape mismatch: a fixed, dated obligation serviced by a variable, correlated income.
Illiquid collateral cannot be sold in pieces, so a shortfall forces an exit rather than a trim.
If the date the money is needed sits inside the asset's characteristic bad patch, the horizon is the problem.
A position whose trigger price cannot be stated in one sentence has not been sized against anything.